cancer_types
ayman.abuelela
13/12/2020
## libraries
library(rgl)
library(TCGAbiolinks)
#library(TCGA2STAT)
library(caret)## Loading required package: lattice## Loading required package: ggplot2library(ggplot2)
library(MASS)
library(heatmap.plus)
library(reshape2)
library(RColorBrewer)
library(ConsensusClusterPlus)
library(sigclust)
library(pheatmap)
library(tsne)
library(gplots)##
## Attaching package: 'gplots'## The following object is masked from 'package:stats':
##
## lowesslibrary(ggradar)
library(doMC)## Loading required package: foreach## Loading required package: iterators## Loading required package: parallellibrary(dplyr)##
## Attaching package: 'dplyr'## The following object is masked from 'package:MASS':
##
## select## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.0 ──## ✓ tibble 3.0.4 ✓ purrr 0.3.4
## ✓ tidyr 1.1.2 ✓ stringr 1.4.0
## ✓ readr 1.4.0 ✓ forcats 0.5.0## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x purrr::accumulate() masks foreach::accumulate()
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
## x purrr::lift() masks caret::lift()
## x dplyr::select() masks MASS::select()
## x purrr::when() masks foreach::when()library(plyr)## ------------------------------------------------------------------------------## You have loaded plyr after dplyr - this is likely to cause problems.
## If you need functions from both plyr and dplyr, please load plyr first, then dplyr:
## library(plyr); library(dplyr)## ------------------------------------------------------------------------------##
## Attaching package: 'plyr'## The following object is masked from 'package:purrr':
##
## compact## The following objects are masked from 'package:dplyr':
##
## arrange, count, desc, failwith, id, mutate, rename, summarise,
## summarizelibrary(MLmetrics)##
## Attaching package: 'MLmetrics'## The following objects are masked from 'package:caret':
##
## MAE, RMSE## The following object is masked from 'package:base':
##
## Recalllibrary(sparcl)# --- importing data ---#
expression_data_dir <- '/Users/mohama32/Documents/projects/GTCancerClassifier/OGFGT_data/'
meta_data_dir <- '/Users/mohama32/Documents/projects/GTCancerClassifier/OGFGT/data'
# import expression data
gt_types_file <- file.path(expression_data_dir, 'Glycosyltransferase_TCGA_RNASeq2_RSEM.txt')
gt_all = read_delim(gt_types_file, col_names = T, delim=" ")##
## ── Column specification ────────────────────────────────────────────────────────
## cols(
## .default = col_double(),
## .id = col_character()
## )
## ℹ Use `spec()` for the full column specifications.# import list of GT genes
gt_file <- file.path(meta_data_dir, 'GT_Genes2.txt')
gt <- read_tsv(gt_file, col_names = FALSE)##
## ── Column specification ────────────────────────────────────────────────────────
## cols(
## X1 = col_character()
## )set.seed(1234)
intrain = createDataPartition(gt_all$.id, p=.7, list=FALSE)
training = gt_all[intrain,]
testing = gt_all[-intrain,]# sample selection
# we may use PCA to better show that certain types can be collapsed in an unsupervised way. I could not do that since the number of types is too many**
lda.fit = train(.id ~ ., data = training, method = "lda")
predictions = predict(lda.fit, testing[,-1])
confusionMatrix(predictions, as.factor(testing$.id))## Confusion Matrix and Statistics
##
## Reference
## Prediction ACC BLCA BRCA CESC CHOL COAD COADREAD DLBC ESCA GBM GBMLGG HNSC KICH
## ACC 20 0 0 0 0 0 0 0 0 0 0 0 0
## BLCA 0 81 4 5 1 0 1 0 0 0 0 0 0
## BRCA 0 0 265 1 0 0 0 0 0 0 0 1 0
## CESC 0 17 3 43 0 0 1 0 2 0 0 12 0
## CHOL 0 0 0 0 3 0 0 0 0 0 0 0 0
## COAD 0 0 0 0 0 19 35 0 1 0 0 1 0
## COADREAD 0 0 0 0 0 46 57 0 0 0 0 0 0
## DLBC 1 0 4 2 0 0 0 9 0 0 0 2 0
## ESCA 0 0 1 0 0 0 0 0 8 0 0 0 0
## GBM 0 0 0 0 0 0 0 0 0 40 66 0 0
## GBMLGG 0 0 0 0 0 0 0 0 0 2 38 0 0
## HNSC 0 13 4 21 0 0 0 0 13 0 0 117 0
## KICH 0 0 0 0 0 0 0 0 0 0 0 0 16
## KIPAN 0 0 0 0 0 0 0 0 0 0 0 0 0
## KIRC 0 0 0 0 0 0 0 0 0 0 0 0 0
## KIRP 0 0 0 0 0 0 0 0 0 0 0 0 0
## LGG 0 0 0 0 0 0 0 0 0 0 95 0 0
## LIHC 0 0 0 1 0 0 0 1 0 0 0 0 0
## LUAD 0 0 2 0 2 0 1 0 0 0 0 0 0
## LUSC 0 6 9 5 0 0 0 0 10 0 0 20 0
## MESO 0 0 0 0 0 0 0 0 0 0 0 1 0
## OV 0 0 1 2 1 0 0 0 1 0 0 0 0
## PAAD 0 0 0 1 2 6 3 0 0 0 0 0 0
## PCPG 0 0 1 0 0 0 0 0 0 0 0 0 0
## PRAD 0 0 9 0 0 0 0 0 0 0 0 0 0
## READ 0 0 0 1 0 9 10 0 1 0 0 0 0
## SARC 1 1 1 0 0 1 0 0 1 2 1 1 0
## SKCM 0 0 2 0 0 0 0 0 0 0 0 0 1
## STAD 0 0 0 0 0 2 5 0 18 0 0 0 0
## TGCT 0 0 0 0 0 0 0 2 0 0 0 0 0
## THCA 0 0 2 0 0 0 0 0 0 0 0 0 0
## THYM 1 1 0 2 1 0 0 1 0 0 0 0 0
## UCEC 0 1 15 7 0 0 0 0 0 0 0 0 0
## UCS 0 2 3 0 0 1 0 1 0 1 0 1 1
## UVM 0 0 1 0 0 0 0 0 0 0 0 0 1
## Reference
## Prediction KIPAN KIRC KIRP LGG LIHC LUAD LUSC MESO OV PAAD PCPG PRAD READ SARC
## ACC 0 0 0 0 0 0 0 0 0 0 0 0 0 1
## BLCA 1 0 1 0 0 2 5 0 0 1 0 0 0 1
## BRCA 0 0 0 0 0 4 2 0 0 0 0 0 0 0
## CESC 1 1 0 0 0 6 11 1 0 1 0 0 0 0
## CHOL 0 0 0 0 0 1 0 0 0 2 0 0 0 0
## COAD 0 0 0 0 0 0 0 0 0 1 0 0 3 0
## COADREAD 0 0 0 0 0 0 0 0 0 1 0 0 15 0
## DLBC 0 0 0 0 0 0 0 0 0 1 0 0 0 0
## ESCA 0 0 0 0 0 0 0 0 0 0 0 1 0 0
## GBM 2 1 0 18 1 0 0 0 0 1 2 0 0 2
## GBMLGG 0 0 0 41 0 0 0 0 0 0 0 0 0 0
## HNSC 0 0 0 0 0 1 8 0 0 0 0 0 0 0
## KICH 26 6 0 0 0 0 0 0 0 0 0 0 0 0
## KIPAN 33 24 25 0 0 0 1 0 0 0 0 0 0 0
## KIRC 128 119 4 0 0 0 0 0 0 0 0 0 0 0
## KIRP 56 3 54 0 0 0 0 0 0 0 0 0 0 0
## LGG 0 0 0 94 0 0 0 0 0 0 0 0 0 0
## LIHC 0 0 0 0 103 0 0 0 0 0 0 0 0 0
## LUAD 1 0 1 0 0 123 6 0 0 1 0 2 0 0
## LUSC 1 0 0 0 0 2 109 0 0 0 0 0 0 1
## MESO 0 0 0 1 0 0 0 20 0 0 0 0 0 2
## OV 1 0 0 0 0 2 4 0 87 0 0 0 0 0
## PAAD 0 0 0 0 1 7 1 0 0 40 0 0 1 0
## PCPG 0 0 0 0 0 0 0 0 0 0 50 0 0 0
## PRAD 0 0 0 0 0 0 0 0 0 0 0 142 0 0
## READ 0 0 0 0 0 0 0 0 0 0 0 0 7 0
## SARC 3 1 1 0 1 1 0 5 0 3 0 3 0 60
## SKCM 1 0 0 0 0 0 0 0 0 0 0 0 0 0
## STAD 0 0 0 0 0 2 0 0 0 0 0 0 2 0
## TGCT 1 0 0 0 0 0 0 0 0 0 0 0 0 0
## THCA 1 0 0 0 0 0 0 0 0 1 0 0 0 3
## THYM 0 1 0 0 0 0 1 0 0 0 0 0 0 1
## UCEC 1 1 0 0 0 2 1 0 1 0 0 1 0 0
## UCS 5 2 1 0 5 1 1 0 2 0 1 0 0 6
## UVM 4 0 0 0 0 0 0 0 0 0 0 0 0 0
## Reference
## Prediction SKCM STAD TGCT THCA THYM UCEC UCS UVM
## ACC 1 0 0 2 0 0 0 0
## BLCA 0 0 0 0 0 0 1 0
## BRCA 0 1 0 1 0 0 0 0
## CESC 0 2 0 0 0 5 0 0
## CHOL 0 0 0 0 0 0 0 0
## COAD 0 1 0 0 0 0 0 0
## COADREAD 0 5 0 0 0 0 0 0
## DLBC 0 0 3 1 5 0 0 0
## ESCA 0 8 0 0 0 0 0 0
## GBM 0 0 0 0 0 0 1 0
## GBMLGG 0 0 0 0 0 0 0 0
## HNSC 3 0 0 0 0 2 0 0
## KICH 0 0 0 0 0 0 0 0
## KIPAN 0 0 0 0 0 0 0 0
## KIRC 0 0 0 0 0 0 0 0
## KIRP 0 0 0 0 0 0 0 0
## LGG 0 1 0 0 0 0 0 0
## LIHC 0 0 0 1 0 0 0 0
## LUAD 0 1 0 3 0 2 0 0
## LUSC 0 1 1 0 0 0 0 0
## MESO 0 0 0 0 0 0 0 0
## OV 0 1 0 0 1 3 1 0
## PAAD 0 1 0 0 0 0 0 0
## PCPG 0 0 0 0 0 0 0 0
## PRAD 0 0 0 0 0 1 0 0
## READ 0 1 0 0 0 0 0 0
## SARC 0 5 0 1 0 0 2 0
## SKCM 19 0 0 0 0 0 0 3
## STAD 0 92 0 0 0 0 0 0
## TGCT 0 3 34 0 0 1 0 0
## THCA 0 0 0 141 0 0 0 0
## THYM 1 0 0 0 30 0 0 0
## UCEC 0 0 0 0 0 37 1 0
## UCS 2 1 2 0 0 1 11 0
## UVM 4 0 0 0 0 0 0 21
##
## Overall Statistics
##
## Accuracy : 0.6513
## 95% CI : (0.6347, 0.6676)
## No Information Rate : 0.0994
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.6365
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: ACC Class: BLCA Class: BRCA Class: CESC Class: CHOL
## Sensitivity 0.869565 0.66393 0.81040 0.47253 0.3000000
## Specificity 0.998775 0.99274 0.99662 0.98030 0.9990851
## Pos Pred Value 0.833333 0.77885 0.96364 0.40566 0.5000000
## Neg Pred Value 0.999081 0.98713 0.97943 0.98492 0.9978678
## Prevalence 0.006993 0.03709 0.09942 0.02767 0.0030404
## Detection Rate 0.006081 0.02463 0.08057 0.01307 0.0009121
## Detection Prevalence 0.007297 0.03162 0.08361 0.03223 0.0018243
## Balanced Accuracy 0.934170 0.82834 0.90351 0.72641 0.6495425
## Class: COAD Class: COADREAD Class: DLBC Class: ESCA
## Sensitivity 0.226190 0.50442 0.642857 0.145455
## Specificity 0.986895 0.97890 0.994198 0.996908
## Pos Pred Value 0.311475 0.45968 0.321429 0.444444
## Neg Pred Value 0.979864 0.98231 0.998467 0.985631
## Prevalence 0.025540 0.03436 0.004257 0.016722
## Detection Rate 0.005777 0.01733 0.002736 0.002432
## Detection Prevalence 0.018547 0.03770 0.008513 0.005473
## Balanced Accuracy 0.606543 0.74166 0.818528 0.571181
## Class: GBM Class: GBMLGG Class: HNSC Class: KICH
## Sensitivity 0.88889 0.19000 0.75000 0.842105
## Specificity 0.97102 0.98608 0.97925 0.990214
## Pos Pred Value 0.29851 0.46914 0.64286 0.333333
## Neg Pred Value 0.99842 0.94950 0.98745 0.999074
## Prevalence 0.01368 0.06081 0.04743 0.005777
## Detection Rate 0.01216 0.01155 0.03557 0.004865
## Detection Prevalence 0.04074 0.02463 0.05534 0.014594
## Balanced Accuracy 0.92996 0.58804 0.86463 0.916160
## Class: KIPAN Class: KIRC Class: KIRP Class: LGG
## Sensitivity 0.12406 0.74843 0.62069 0.61039
## Specificity 0.98346 0.95783 0.98157 0.96938
## Pos Pred Value 0.39759 0.47410 0.47788 0.49474
## Neg Pred Value 0.92732 0.98683 0.98961 0.98064
## Prevalence 0.08088 0.04834 0.02645 0.04682
## Detection Rate 0.01003 0.03618 0.01642 0.02858
## Detection Prevalence 0.02524 0.07631 0.03436 0.05777
## Balanced Accuracy 0.55376 0.85313 0.80113 0.78988
## Class: LIHC Class: LUAD Class: LUSC Class: MESO Class: OV
## Sensitivity 0.92793 0.79870 0.72667 0.769231 0.96667
## Specificity 0.99906 0.99298 0.98216 0.998774 0.99437
## Pos Pred Value 0.97170 0.84828 0.66061 0.833333 0.82857
## Neg Pred Value 0.99749 0.99014 0.98688 0.998162 0.99906
## Prevalence 0.03375 0.04682 0.04561 0.007905 0.02736
## Detection Rate 0.03132 0.03740 0.03314 0.006081 0.02645
## Detection Prevalence 0.03223 0.04409 0.05017 0.007297 0.03192
## Balanced Accuracy 0.96349 0.89584 0.85441 0.884002 0.98052
## Class: PAAD Class: PCPG Class: PRAD Class: READ
## Sensitivity 0.75472 0.94340 0.95302 0.250000
## Specificity 0.99289 0.99969 0.99682 0.993254
## Pos Pred Value 0.63492 0.98039 0.93421 0.241379
## Neg Pred Value 0.99597 0.99907 0.99777 0.993558
## Prevalence 0.01611 0.01611 0.04530 0.008513
## Detection Rate 0.01216 0.01520 0.04317 0.002128
## Detection Prevalence 0.01915 0.01551 0.04621 0.008817
## Balanced Accuracy 0.87380 0.97154 0.97492 0.621627
## Class: SARC Class: SKCM Class: STAD Class: TGCT
## Sensitivity 0.77922 0.633333 0.74194 0.85000
## Specificity 0.98910 0.997852 0.99084 0.99785
## Pos Pred Value 0.63158 0.730769 0.76033 0.82927
## Neg Pred Value 0.99468 0.996629 0.98990 0.99815
## Prevalence 0.02341 0.009121 0.03770 0.01216
## Detection Rate 0.01824 0.005777 0.02797 0.01034
## Detection Prevalence 0.02888 0.007905 0.03679 0.01247
## Balanced Accuracy 0.88416 0.815593 0.86639 0.92392
## Class: THCA Class: THYM Class: UCEC Class: UCS Class: UVM
## Sensitivity 0.94000 0.833333 0.71154 0.647059 0.875000
## Specificity 0.99777 0.996926 0.99042 0.987775 0.996937
## Pos Pred Value 0.95270 0.750000 0.54412 0.215686 0.677419
## Neg Pred Value 0.99713 0.998153 0.99534 0.998147 0.999079
## Prevalence 0.04561 0.010946 0.01581 0.005169 0.007297
## Detection Rate 0.04287 0.009121 0.01125 0.003344 0.006385
## Detection Prevalence 0.04500 0.012162 0.02067 0.015506 0.009425
## Balanced Accuracy 0.96888 0.915130 0.85098 0.817417 0.935969pheatmap(table(predictions, testing$.id), scale = "column", cluster_rows = FALSE, cluster_cols=FALSE)
# --- collapsing samples by type ---#
mod_id <- gt_all$.id
mod_id <- gsub("^COAD$|^COADREAD$|^READ$", "COLORECTAL", mod_id)
mod_id <- gsub("^LUAD$|^LUSC$", "LUNG", mod_id)
mod_id <- gsub("^UCS$|^UCEC$", "UTERINE", mod_id)
mod_id <- gsub("^PAAD$|^LIHC$|^CHOL$", "PARALIVER", mod_id)
mod_id <- gsub("^ESCA$|^STAD$", "STOPH", mod_id)
mod_id <- gsub("^GBM$|^GBMLGG$|^LGG$", "GLIOMA", mod_id)
mod_id <- gsub("^KICH$|^KIPAN$|^KIRP$|^KIRC$","KIDNEY", mod_id)
gt_all2 = cbind(mod_id, gt_all[,-1])
unique(gt_all2$mod_id)## [1] "ACC" "BLCA" "BRCA" "CESC" "PARALIVER"
## [6] "COLORECTAL" "DLBC" "STOPH" "GLIOMA" "HNSC"
## [11] "KIDNEY" "LUNG" "MESO" "OV" "PCPG"
## [16] "PRAD" "SARC" "SKCM" "TGCT" "THCA"
## [21] "THYM" "UTERINE" "UVM"confusionMatrix(predictions, as.factor(testing$.id))## Confusion Matrix and Statistics
##
## Reference
## Prediction ACC BLCA BRCA CESC CHOL COAD COADREAD DLBC ESCA GBM GBMLGG HNSC KICH
## ACC 20 0 0 0 0 0 0 0 0 0 0 0 0
## BLCA 0 81 4 5 1 0 1 0 0 0 0 0 0
## BRCA 0 0 265 1 0 0 0 0 0 0 0 1 0
## CESC 0 17 3 43 0 0 1 0 2 0 0 12 0
## CHOL 0 0 0 0 3 0 0 0 0 0 0 0 0
## COAD 0 0 0 0 0 19 35 0 1 0 0 1 0
## COADREAD 0 0 0 0 0 46 57 0 0 0 0 0 0
## DLBC 1 0 4 2 0 0 0 9 0 0 0 2 0
## ESCA 0 0 1 0 0 0 0 0 8 0 0 0 0
## GBM 0 0 0 0 0 0 0 0 0 40 66 0 0
## GBMLGG 0 0 0 0 0 0 0 0 0 2 38 0 0
## HNSC 0 13 4 21 0 0 0 0 13 0 0 117 0
## KICH 0 0 0 0 0 0 0 0 0 0 0 0 16
## KIPAN 0 0 0 0 0 0 0 0 0 0 0 0 0
## KIRC 0 0 0 0 0 0 0 0 0 0 0 0 0
## KIRP 0 0 0 0 0 0 0 0 0 0 0 0 0
## LGG 0 0 0 0 0 0 0 0 0 0 95 0 0
## LIHC 0 0 0 1 0 0 0 1 0 0 0 0 0
## LUAD 0 0 2 0 2 0 1 0 0 0 0 0 0
## LUSC 0 6 9 5 0 0 0 0 10 0 0 20 0
## MESO 0 0 0 0 0 0 0 0 0 0 0 1 0
## OV 0 0 1 2 1 0 0 0 1 0 0 0 0
## PAAD 0 0 0 1 2 6 3 0 0 0 0 0 0
## PCPG 0 0 1 0 0 0 0 0 0 0 0 0 0
## PRAD 0 0 9 0 0 0 0 0 0 0 0 0 0
## READ 0 0 0 1 0 9 10 0 1 0 0 0 0
## SARC 1 1 1 0 0 1 0 0 1 2 1 1 0
## SKCM 0 0 2 0 0 0 0 0 0 0 0 0 1
## STAD 0 0 0 0 0 2 5 0 18 0 0 0 0
## TGCT 0 0 0 0 0 0 0 2 0 0 0 0 0
## THCA 0 0 2 0 0 0 0 0 0 0 0 0 0
## THYM 1 1 0 2 1 0 0 1 0 0 0 0 0
## UCEC 0 1 15 7 0 0 0 0 0 0 0 0 0
## UCS 0 2 3 0 0 1 0 1 0 1 0 1 1
## UVM 0 0 1 0 0 0 0 0 0 0 0 0 1
## Reference
## Prediction KIPAN KIRC KIRP LGG LIHC LUAD LUSC MESO OV PAAD PCPG PRAD READ SARC
## ACC 0 0 0 0 0 0 0 0 0 0 0 0 0 1
## BLCA 1 0 1 0 0 2 5 0 0 1 0 0 0 1
## BRCA 0 0 0 0 0 4 2 0 0 0 0 0 0 0
## CESC 1 1 0 0 0 6 11 1 0 1 0 0 0 0
## CHOL 0 0 0 0 0 1 0 0 0 2 0 0 0 0
## COAD 0 0 0 0 0 0 0 0 0 1 0 0 3 0
## COADREAD 0 0 0 0 0 0 0 0 0 1 0 0 15 0
## DLBC 0 0 0 0 0 0 0 0 0 1 0 0 0 0
## ESCA 0 0 0 0 0 0 0 0 0 0 0 1 0 0
## GBM 2 1 0 18 1 0 0 0 0 1 2 0 0 2
## GBMLGG 0 0 0 41 0 0 0 0 0 0 0 0 0 0
## HNSC 0 0 0 0 0 1 8 0 0 0 0 0 0 0
## KICH 26 6 0 0 0 0 0 0 0 0 0 0 0 0
## KIPAN 33 24 25 0 0 0 1 0 0 0 0 0 0 0
## KIRC 128 119 4 0 0 0 0 0 0 0 0 0 0 0
## KIRP 56 3 54 0 0 0 0 0 0 0 0 0 0 0
## LGG 0 0 0 94 0 0 0 0 0 0 0 0 0 0
## LIHC 0 0 0 0 103 0 0 0 0 0 0 0 0 0
## LUAD 1 0 1 0 0 123 6 0 0 1 0 2 0 0
## LUSC 1 0 0 0 0 2 109 0 0 0 0 0 0 1
## MESO 0 0 0 1 0 0 0 20 0 0 0 0 0 2
## OV 1 0 0 0 0 2 4 0 87 0 0 0 0 0
## PAAD 0 0 0 0 1 7 1 0 0 40 0 0 1 0
## PCPG 0 0 0 0 0 0 0 0 0 0 50 0 0 0
## PRAD 0 0 0 0 0 0 0 0 0 0 0 142 0 0
## READ 0 0 0 0 0 0 0 0 0 0 0 0 7 0
## SARC 3 1 1 0 1 1 0 5 0 3 0 3 0 60
## SKCM 1 0 0 0 0 0 0 0 0 0 0 0 0 0
## STAD 0 0 0 0 0 2 0 0 0 0 0 0 2 0
## TGCT 1 0 0 0 0 0 0 0 0 0 0 0 0 0
## THCA 1 0 0 0 0 0 0 0 0 1 0 0 0 3
## THYM 0 1 0 0 0 0 1 0 0 0 0 0 0 1
## UCEC 1 1 0 0 0 2 1 0 1 0 0 1 0 0
## UCS 5 2 1 0 5 1 1 0 2 0 1 0 0 6
## UVM 4 0 0 0 0 0 0 0 0 0 0 0 0 0
## Reference
## Prediction SKCM STAD TGCT THCA THYM UCEC UCS UVM
## ACC 1 0 0 2 0 0 0 0
## BLCA 0 0 0 0 0 0 1 0
## BRCA 0 1 0 1 0 0 0 0
## CESC 0 2 0 0 0 5 0 0
## CHOL 0 0 0 0 0 0 0 0
## COAD 0 1 0 0 0 0 0 0
## COADREAD 0 5 0 0 0 0 0 0
## DLBC 0 0 3 1 5 0 0 0
## ESCA 0 8 0 0 0 0 0 0
## GBM 0 0 0 0 0 0 1 0
## GBMLGG 0 0 0 0 0 0 0 0
## HNSC 3 0 0 0 0 2 0 0
## KICH 0 0 0 0 0 0 0 0
## KIPAN 0 0 0 0 0 0 0 0
## KIRC 0 0 0 0 0 0 0 0
## KIRP 0 0 0 0 0 0 0 0
## LGG 0 1 0 0 0 0 0 0
## LIHC 0 0 0 1 0 0 0 0
## LUAD 0 1 0 3 0 2 0 0
## LUSC 0 1 1 0 0 0 0 0
## MESO 0 0 0 0 0 0 0 0
## OV 0 1 0 0 1 3 1 0
## PAAD 0 1 0 0 0 0 0 0
## PCPG 0 0 0 0 0 0 0 0
## PRAD 0 0 0 0 0 1 0 0
## READ 0 1 0 0 0 0 0 0
## SARC 0 5 0 1 0 0 2 0
## SKCM 19 0 0 0 0 0 0 3
## STAD 0 92 0 0 0 0 0 0
## TGCT 0 3 34 0 0 1 0 0
## THCA 0 0 0 141 0 0 0 0
## THYM 1 0 0 0 30 0 0 0
## UCEC 0 0 0 0 0 37 1 0
## UCS 2 1 2 0 0 1 11 0
## UVM 4 0 0 0 0 0 0 21
##
## Overall Statistics
##
## Accuracy : 0.6513
## 95% CI : (0.6347, 0.6676)
## No Information Rate : 0.0994
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.6365
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: ACC Class: BLCA Class: BRCA Class: CESC Class: CHOL
## Sensitivity 0.869565 0.66393 0.81040 0.47253 0.3000000
## Specificity 0.998775 0.99274 0.99662 0.98030 0.9990851
## Pos Pred Value 0.833333 0.77885 0.96364 0.40566 0.5000000
## Neg Pred Value 0.999081 0.98713 0.97943 0.98492 0.9978678
## Prevalence 0.006993 0.03709 0.09942 0.02767 0.0030404
## Detection Rate 0.006081 0.02463 0.08057 0.01307 0.0009121
## Detection Prevalence 0.007297 0.03162 0.08361 0.03223 0.0018243
## Balanced Accuracy 0.934170 0.82834 0.90351 0.72641 0.6495425
## Class: COAD Class: COADREAD Class: DLBC Class: ESCA
## Sensitivity 0.226190 0.50442 0.642857 0.145455
## Specificity 0.986895 0.97890 0.994198 0.996908
## Pos Pred Value 0.311475 0.45968 0.321429 0.444444
## Neg Pred Value 0.979864 0.98231 0.998467 0.985631
## Prevalence 0.025540 0.03436 0.004257 0.016722
## Detection Rate 0.005777 0.01733 0.002736 0.002432
## Detection Prevalence 0.018547 0.03770 0.008513 0.005473
## Balanced Accuracy 0.606543 0.74166 0.818528 0.571181
## Class: GBM Class: GBMLGG Class: HNSC Class: KICH
## Sensitivity 0.88889 0.19000 0.75000 0.842105
## Specificity 0.97102 0.98608 0.97925 0.990214
## Pos Pred Value 0.29851 0.46914 0.64286 0.333333
## Neg Pred Value 0.99842 0.94950 0.98745 0.999074
## Prevalence 0.01368 0.06081 0.04743 0.005777
## Detection Rate 0.01216 0.01155 0.03557 0.004865
## Detection Prevalence 0.04074 0.02463 0.05534 0.014594
## Balanced Accuracy 0.92996 0.58804 0.86463 0.916160
## Class: KIPAN Class: KIRC Class: KIRP Class: LGG
## Sensitivity 0.12406 0.74843 0.62069 0.61039
## Specificity 0.98346 0.95783 0.98157 0.96938
## Pos Pred Value 0.39759 0.47410 0.47788 0.49474
## Neg Pred Value 0.92732 0.98683 0.98961 0.98064
## Prevalence 0.08088 0.04834 0.02645 0.04682
## Detection Rate 0.01003 0.03618 0.01642 0.02858
## Detection Prevalence 0.02524 0.07631 0.03436 0.05777
## Balanced Accuracy 0.55376 0.85313 0.80113 0.78988
## Class: LIHC Class: LUAD Class: LUSC Class: MESO Class: OV
## Sensitivity 0.92793 0.79870 0.72667 0.769231 0.96667
## Specificity 0.99906 0.99298 0.98216 0.998774 0.99437
## Pos Pred Value 0.97170 0.84828 0.66061 0.833333 0.82857
## Neg Pred Value 0.99749 0.99014 0.98688 0.998162 0.99906
## Prevalence 0.03375 0.04682 0.04561 0.007905 0.02736
## Detection Rate 0.03132 0.03740 0.03314 0.006081 0.02645
## Detection Prevalence 0.03223 0.04409 0.05017 0.007297 0.03192
## Balanced Accuracy 0.96349 0.89584 0.85441 0.884002 0.98052
## Class: PAAD Class: PCPG Class: PRAD Class: READ
## Sensitivity 0.75472 0.94340 0.95302 0.250000
## Specificity 0.99289 0.99969 0.99682 0.993254
## Pos Pred Value 0.63492 0.98039 0.93421 0.241379
## Neg Pred Value 0.99597 0.99907 0.99777 0.993558
## Prevalence 0.01611 0.01611 0.04530 0.008513
## Detection Rate 0.01216 0.01520 0.04317 0.002128
## Detection Prevalence 0.01915 0.01551 0.04621 0.008817
## Balanced Accuracy 0.87380 0.97154 0.97492 0.621627
## Class: SARC Class: SKCM Class: STAD Class: TGCT
## Sensitivity 0.77922 0.633333 0.74194 0.85000
## Specificity 0.98910 0.997852 0.99084 0.99785
## Pos Pred Value 0.63158 0.730769 0.76033 0.82927
## Neg Pred Value 0.99468 0.996629 0.98990 0.99815
## Prevalence 0.02341 0.009121 0.03770 0.01216
## Detection Rate 0.01824 0.005777 0.02797 0.01034
## Detection Prevalence 0.02888 0.007905 0.03679 0.01247
## Balanced Accuracy 0.88416 0.815593 0.86639 0.92392
## Class: THCA Class: THYM Class: UCEC Class: UCS Class: UVM
## Sensitivity 0.94000 0.833333 0.71154 0.647059 0.875000
## Specificity 0.99777 0.996926 0.99042 0.987775 0.996937
## Pos Pred Value 0.95270 0.750000 0.54412 0.215686 0.677419
## Neg Pred Value 0.99713 0.998153 0.99534 0.998147 0.999079
## Prevalence 0.04561 0.010946 0.01581 0.005169 0.007297
## Detection Rate 0.04287 0.009121 0.01125 0.003344 0.006385
## Detection Prevalence 0.04500 0.012162 0.02067 0.015506 0.009425
## Balanced Accuracy 0.96888 0.915130 0.85098 0.817417 0.935969# --- preprocessing ---#
set.seed(1234)
intrain = createDataPartition(y = mod_id, p = 0.7, list = FALSE)
training = gt_all2[intrain,]
testing = gt_all2[-intrain,]
set.seed(1234)
pp = preProcess(training, method = c("nzv", "scale", "center", "YeoJohnson"))
pptraining = predict(pp, training)
pptesting = predict(pp, testing)# --- I think this should stay to show that rda was picked based on performance, no bias ---#
# note - glioma among best performing
# --- algorithm/method choice ---#
set.seed(123456)
#methods = c("pam", "knn", "rf", "xgbTree", "lda", "rda", "xgbTree", "lda")
methods = c("rda")
measures = vector(length = length(methods), mode = "list")
for (i in 1:length(methods)) {
method = methods[i]
print(method)
model.fit = train(mod_id ~ ., data = pptraining, method = method)
predictions = predict(model.fit, pptesting[,-1])
confmat = confusionMatrix(predictions, as.factor(pptesting$mod_id))
print(confmat)
pheatmap(table(predictions, pptesting$mod_id), scale = "column", cluster_rows=FALSE, cluster_cols=FALSE)
names(measures)[i] = method
measures[[i]] = confmat$byClass
}## [1] "rda"
## Confusion Matrix and Statistics
##
## Reference
## Prediction ACC BLCA BRCA CESC COLORECTAL DLBC GLIOMA HNSC KIDNEY LUNG MESO
## ACC 22 0 0 0 0 0 0 0 0 0 0
## BLCA 0 100 2 2 0 0 0 1 3 4 0
## BRCA 0 1 319 0 0 0 0 0 0 0 0
## CESC 0 3 0 71 2 0 0 4 0 6 0
## COLORECTAL 0 0 0 1 221 0 0 0 0 0 0
## DLBC 0 0 0 0 0 14 0 0 0 0 0
## GLIOMA 0 0 0 0 0 0 398 0 0 0 0
## HNSC 0 8 2 8 0 0 0 134 0 20 0
## KIDNEY 0 0 0 0 0 0 0 0 528 0 0
## LUNG 0 6 1 3 0 0 0 14 0 262 0
## MESO 0 0 0 0 0 0 0 0 0 0 25
## OV 0 0 0 0 0 0 0 0 0 2 0
## PARALIVER 0 0 0 0 0 0 0 0 0 1 0
## PCPG 0 0 0 0 0 0 0 0 0 1 0
## PRAD 0 0 0 0 0 0 0 0 0 0 0
## SARC 1 2 3 0 0 0 2 2 2 1 1
## SKCM 0 0 0 0 0 0 0 0 0 0 0
## STOPH 0 1 0 2 3 0 0 0 0 5 0
## TGCT 0 0 0 0 0 0 0 0 0 0 0
## THCA 0 0 0 0 0 0 0 0 0 0 0
## THYM 0 0 0 0 0 0 0 0 0 0 0
## UTERINE 0 1 0 4 0 0 0 1 0 2 0
## UVM 0 0 0 0 0 0 0 0 0 0 0
## Reference
## Prediction OV PARALIVER PCPG PRAD SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## ACC 0 0 0 0 0 0 0 0 0 0 1 0
## BLCA 0 0 0 0 0 0 0 0 0 0 0 0
## BRCA 0 0 0 0 0 0 0 0 0 0 0 0
## CESC 0 0 0 0 0 0 0 0 0 0 0 0
## COLORECTAL 0 3 0 0 0 0 6 0 0 0 0 0
## DLBC 0 0 0 0 0 0 1 0 0 0 0 0
## GLIOMA 0 0 0 0 0 0 0 0 0 0 0 0
## HNSC 0 0 0 0 0 0 9 0 0 1 0 0
## KIDNEY 0 1 0 0 0 0 0 0 0 0 0 0
## LUNG 0 2 0 0 0 0 7 0 0 0 1 0
## MESO 0 0 0 0 0 0 0 0 0 0 0 0
## OV 85 0 0 0 0 0 0 0 0 0 4 0
## PARALIVER 0 165 0 0 0 0 0 1 0 0 0 0
## PCPG 0 0 53 0 0 0 0 0 0 0 0 0
## PRAD 0 0 0 149 0 0 0 0 0 0 0 0
## SARC 1 3 0 0 75 0 0 0 0 0 3 0
## SKCM 0 0 0 0 1 29 0 0 0 0 0 0
## STOPH 0 1 0 0 0 0 156 0 0 0 0 0
## TGCT 0 0 0 0 0 0 0 38 0 0 0 0
## THCA 0 0 0 0 0 0 0 0 150 0 0 0
## THYM 0 0 0 0 0 0 0 0 0 34 0 0
## UTERINE 4 0 0 0 1 0 0 1 0 1 60 0
## UVM 0 0 0 0 0 1 0 0 0 0 0 24
##
## Overall Statistics
##
## Accuracy : 0.9447
## 95% CI : (0.9364, 0.9523)
## No Information Rate : 0.1618
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9399
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: ACC Class: BLCA Class: BRCA Class: CESC
## Sensitivity 0.956522 0.81967 0.97554 0.78022
## Specificity 0.999694 0.99622 0.99966 0.99532
## Pos Pred Value 0.956522 0.89286 0.99687 0.82558
## Neg Pred Value 0.999694 0.99309 0.99731 0.99377
## Prevalence 0.006982 0.03704 0.09927 0.02763
## Detection Rate 0.006679 0.03036 0.09684 0.02155
## Detection Prevalence 0.006982 0.03400 0.09715 0.02611
## Balanced Accuracy 0.978108 0.90794 0.98760 0.88777
## Class: COLORECTAL Class: DLBC Class: GLIOMA Class: HNSC
## Sensitivity 0.97788 1.000000 0.9950 0.85897
## Specificity 0.99674 0.999695 1.0000 0.98470
## Pos Pred Value 0.95671 0.933333 1.0000 0.73626
## Neg Pred Value 0.99837 1.000000 0.9993 0.99293
## Prevalence 0.06861 0.004250 0.1214 0.04736
## Detection Rate 0.06709 0.004250 0.1208 0.04068
## Detection Prevalence 0.07013 0.004554 0.1208 0.05525
## Balanced Accuracy 0.98731 0.999848 0.9975 0.92184
## Class: KIDNEY Class: LUNG Class: MESO Class: OV
## Sensitivity 0.9906 0.86184 0.961538 0.94444
## Specificity 0.9996 0.98863 1.000000 0.99813
## Pos Pred Value 0.9981 0.88514 1.000000 0.93407
## Neg Pred Value 0.9982 0.98599 0.999694 0.99844
## Prevalence 0.1618 0.09229 0.007893 0.02732
## Detection Rate 0.1603 0.07954 0.007590 0.02580
## Detection Prevalence 0.1606 0.08986 0.007590 0.02763
## Balanced Accuracy 0.9951 0.92524 0.980769 0.97129
## Class: PARALIVER Class: PCPG Class: PRAD Class: SARC
## Sensitivity 0.94286 1.00000 1.00000 0.97403
## Specificity 0.99936 0.99969 1.00000 0.99347
## Pos Pred Value 0.98802 0.98148 1.00000 0.78125
## Neg Pred Value 0.99680 1.00000 1.00000 0.99937
## Prevalence 0.05313 0.01609 0.04523 0.02338
## Detection Rate 0.05009 0.01609 0.04523 0.02277
## Detection Prevalence 0.05070 0.01639 0.04523 0.02914
## Balanced Accuracy 0.97111 0.99985 1.00000 0.98375
## Class: SKCM Class: STOPH Class: TGCT Class: THCA
## Sensitivity 0.966667 0.87151 0.95000 1.00000
## Specificity 0.999694 0.99615 1.00000 1.00000
## Pos Pred Value 0.966667 0.92857 1.00000 1.00000
## Neg Pred Value 0.999694 0.99264 0.99939 1.00000
## Prevalence 0.009107 0.05434 0.01214 0.04554
## Detection Rate 0.008804 0.04736 0.01154 0.04554
## Detection Prevalence 0.009107 0.05100 0.01154 0.04554
## Balanced Accuracy 0.983180 0.93383 0.97500 1.00000
## Class: THYM Class: UTERINE Class: UVM
## Sensitivity 0.94444 0.86957 1.000000
## Specificity 1.00000 0.99535 0.999694
## Pos Pred Value 1.00000 0.80000 0.960000
## Neg Pred Value 0.99939 0.99720 1.000000
## Prevalence 0.01093 0.02095 0.007286
## Detection Rate 0.01032 0.01821 0.007286
## Detection Prevalence 0.01032 0.02277 0.007590
## Balanced Accuracy 0.97222 0.93246 0.999847
print(confmat)## Confusion Matrix and Statistics
##
## Reference
## Prediction ACC BLCA BRCA CESC COLORECTAL DLBC GLIOMA HNSC KIDNEY LUNG MESO
## ACC 22 0 0 0 0 0 0 0 0 0 0
## BLCA 0 100 2 2 0 0 0 1 3 4 0
## BRCA 0 1 319 0 0 0 0 0 0 0 0
## CESC 0 3 0 71 2 0 0 4 0 6 0
## COLORECTAL 0 0 0 1 221 0 0 0 0 0 0
## DLBC 0 0 0 0 0 14 0 0 0 0 0
## GLIOMA 0 0 0 0 0 0 398 0 0 0 0
## HNSC 0 8 2 8 0 0 0 134 0 20 0
## KIDNEY 0 0 0 0 0 0 0 0 528 0 0
## LUNG 0 6 1 3 0 0 0 14 0 262 0
## MESO 0 0 0 0 0 0 0 0 0 0 25
## OV 0 0 0 0 0 0 0 0 0 2 0
## PARALIVER 0 0 0 0 0 0 0 0 0 1 0
## PCPG 0 0 0 0 0 0 0 0 0 1 0
## PRAD 0 0 0 0 0 0 0 0 0 0 0
## SARC 1 2 3 0 0 0 2 2 2 1 1
## SKCM 0 0 0 0 0 0 0 0 0 0 0
## STOPH 0 1 0 2 3 0 0 0 0 5 0
## TGCT 0 0 0 0 0 0 0 0 0 0 0
## THCA 0 0 0 0 0 0 0 0 0 0 0
## THYM 0 0 0 0 0 0 0 0 0 0 0
## UTERINE 0 1 0 4 0 0 0 1 0 2 0
## UVM 0 0 0 0 0 0 0 0 0 0 0
## Reference
## Prediction OV PARALIVER PCPG PRAD SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## ACC 0 0 0 0 0 0 0 0 0 0 1 0
## BLCA 0 0 0 0 0 0 0 0 0 0 0 0
## BRCA 0 0 0 0 0 0 0 0 0 0 0 0
## CESC 0 0 0 0 0 0 0 0 0 0 0 0
## COLORECTAL 0 3 0 0 0 0 6 0 0 0 0 0
## DLBC 0 0 0 0 0 0 1 0 0 0 0 0
## GLIOMA 0 0 0 0 0 0 0 0 0 0 0 0
## HNSC 0 0 0 0 0 0 9 0 0 1 0 0
## KIDNEY 0 1 0 0 0 0 0 0 0 0 0 0
## LUNG 0 2 0 0 0 0 7 0 0 0 1 0
## MESO 0 0 0 0 0 0 0 0 0 0 0 0
## OV 85 0 0 0 0 0 0 0 0 0 4 0
## PARALIVER 0 165 0 0 0 0 0 1 0 0 0 0
## PCPG 0 0 53 0 0 0 0 0 0 0 0 0
## PRAD 0 0 0 149 0 0 0 0 0 0 0 0
## SARC 1 3 0 0 75 0 0 0 0 0 3 0
## SKCM 0 0 0 0 1 29 0 0 0 0 0 0
## STOPH 0 1 0 0 0 0 156 0 0 0 0 0
## TGCT 0 0 0 0 0 0 0 38 0 0 0 0
## THCA 0 0 0 0 0 0 0 0 150 0 0 0
## THYM 0 0 0 0 0 0 0 0 0 34 0 0
## UTERINE 4 0 0 0 1 0 0 1 0 1 60 0
## UVM 0 0 0 0 0 1 0 0 0 0 0 24
##
## Overall Statistics
##
## Accuracy : 0.9447
## 95% CI : (0.9364, 0.9523)
## No Information Rate : 0.1618
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.9399
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: ACC Class: BLCA Class: BRCA Class: CESC
## Sensitivity 0.956522 0.81967 0.97554 0.78022
## Specificity 0.999694 0.99622 0.99966 0.99532
## Pos Pred Value 0.956522 0.89286 0.99687 0.82558
## Neg Pred Value 0.999694 0.99309 0.99731 0.99377
## Prevalence 0.006982 0.03704 0.09927 0.02763
## Detection Rate 0.006679 0.03036 0.09684 0.02155
## Detection Prevalence 0.006982 0.03400 0.09715 0.02611
## Balanced Accuracy 0.978108 0.90794 0.98760 0.88777
## Class: COLORECTAL Class: DLBC Class: GLIOMA Class: HNSC
## Sensitivity 0.97788 1.000000 0.9950 0.85897
## Specificity 0.99674 0.999695 1.0000 0.98470
## Pos Pred Value 0.95671 0.933333 1.0000 0.73626
## Neg Pred Value 0.99837 1.000000 0.9993 0.99293
## Prevalence 0.06861 0.004250 0.1214 0.04736
## Detection Rate 0.06709 0.004250 0.1208 0.04068
## Detection Prevalence 0.07013 0.004554 0.1208 0.05525
## Balanced Accuracy 0.98731 0.999848 0.9975 0.92184
## Class: KIDNEY Class: LUNG Class: MESO Class: OV
## Sensitivity 0.9906 0.86184 0.961538 0.94444
## Specificity 0.9996 0.98863 1.000000 0.99813
## Pos Pred Value 0.9981 0.88514 1.000000 0.93407
## Neg Pred Value 0.9982 0.98599 0.999694 0.99844
## Prevalence 0.1618 0.09229 0.007893 0.02732
## Detection Rate 0.1603 0.07954 0.007590 0.02580
## Detection Prevalence 0.1606 0.08986 0.007590 0.02763
## Balanced Accuracy 0.9951 0.92524 0.980769 0.97129
## Class: PARALIVER Class: PCPG Class: PRAD Class: SARC
## Sensitivity 0.94286 1.00000 1.00000 0.97403
## Specificity 0.99936 0.99969 1.00000 0.99347
## Pos Pred Value 0.98802 0.98148 1.00000 0.78125
## Neg Pred Value 0.99680 1.00000 1.00000 0.99937
## Prevalence 0.05313 0.01609 0.04523 0.02338
## Detection Rate 0.05009 0.01609 0.04523 0.02277
## Detection Prevalence 0.05070 0.01639 0.04523 0.02914
## Balanced Accuracy 0.97111 0.99985 1.00000 0.98375
## Class: SKCM Class: STOPH Class: TGCT Class: THCA
## Sensitivity 0.966667 0.87151 0.95000 1.00000
## Specificity 0.999694 0.99615 1.00000 1.00000
## Pos Pred Value 0.966667 0.92857 1.00000 1.00000
## Neg Pred Value 0.999694 0.99264 0.99939 1.00000
## Prevalence 0.009107 0.05434 0.01214 0.04554
## Detection Rate 0.008804 0.04736 0.01154 0.04554
## Detection Prevalence 0.009107 0.05100 0.01154 0.04554
## Balanced Accuracy 0.983180 0.93383 0.97500 1.00000
## Class: THYM Class: UTERINE Class: UVM
## Sensitivity 0.94444 0.86957 1.000000
## Specificity 1.00000 0.99535 0.999694
## Pos Pred Value 1.00000 0.80000 0.960000
## Neg Pred Value 0.99939 0.99720 1.000000
## Prevalence 0.01093 0.02095 0.007286
## Detection Rate 0.01032 0.01821 0.007286
## Detection Prevalence 0.01032 0.02277 0.007590
## Balanced Accuracy 0.97222 0.93246 0.999847measuresDF = ldply(measures, data.frame)# --- method comparison ---#
type = unique(mod_id)
measuresDF$type = rep(type, length(unique(measuresDF$.id)))
measuresDFm = melt(measuresDF)
measuresDFm = dcast(measuresDFm, .id + variable ~ type)
measuresDFm = measuresDFm[-grep("Prevalence|Detection", measuresDFm$variable),]
measuresList = split(measuresDFm, as.character(measuresDFm$variable))
for (i in 1:length(measuresList)) {
print(names(measuresList)[i])
print(measuresList[[i]])
print(ggradar(measuresList[[i]][,-2]))
}## [1] "Balanced.Accuracy"
## .id variable ACC BLCA BRCA CESC COLORECTAL
## 11 rda Balanced.Accuracy 0.978108 0.9079445 0.9875991 0.8877683 0.9998476
## DLBC GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER
## 11 0.9975 0.9951285 0.9252354 0.9807692 0.9712859 0.971108 0.9998457 0.9873083
## PCPG PRAD SARC SKCM STOPH TGCT THCA THYM UTERINE
## 11 1 0.9837491 0.9831801 0.933828 0.921839 0.975 1 0.9722222 0.932457
## UVM
## 11 0.9998471
## [1] "F1"
## .id variable ACC BLCA BRCA CESC COLORECTAL DLBC
## 7 rda F1 0.9565217 0.8547009 0.9860896 0.8022599 0.9655172 0.9974937
## GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER PCPG
## 7 0.9943503 0.8733333 0.9803922 0.9392265 0.9649123 0.9906542 0.9671772 1
## PRAD SARC SKCM STOPH TGCT THCA THYM UTERINE
## 7 0.867052 0.9666667 0.8991354 0.7928994 0.974359 1 0.9714286 0.8333333
## UVM
## 7 0.9795918
## [1] "Neg.Pred.Value"
## .id variable ACC BLCA BRCA CESC COLORECTAL DLBC
## 4 rda Neg.Pred.Value 0.9996943 0.9930861 0.99731 0.9937656 1 0.9993094
## GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER PCPG PRAD
## 4 0.9981917 0.9859907 0.9996941 0.998439 0.996802 1 0.9983676 1 0.9993746
## SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## 4 0.9996936 0.9926424 0.9929306 0.9993857 1 0.9993865 0.9972041 1
## [1] "Pos.Pred.Value"
## .id variable ACC BLCA BRCA CESC COLORECTAL DLBC
## 3 rda Pos.Pred.Value 0.9565217 0.8928571 0.996875 0.8255814 0.9333333 1
## GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER PCPG
## 3 0.9981096 0.8851351 1 0.9340659 0.988024 0.9814815 0.95671 1
## PRAD SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## 3 0.78125 0.9666667 0.9285714 0.7362637 1 1 1 0.8 0.96
## [1] "Precision"
## .id variable ACC BLCA BRCA CESC COLORECTAL DLBC
## 5 rda Precision 0.9565217 0.8928571 0.996875 0.8255814 0.9333333 1
## GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER PCPG
## 5 0.9981096 0.8851351 1 0.9340659 0.988024 0.9814815 0.95671 1
## PRAD SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## 5 0.78125 0.9666667 0.9285714 0.7362637 1 1 1 0.8 0.96
## [1] "Recall"
## .id variable ACC BLCA BRCA CESC COLORECTAL DLBC
## 6 rda Recall 0.9565217 0.8196721 0.9755352 0.7802198 1 0.995
## GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER PCPG PRAD
## 6 0.9906191 0.8618421 0.9615385 0.9444444 0.9428571 1 0.9778761 1 0.974026
## SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## 6 0.9666667 0.8715084 0.8589744 0.95 1 0.9444444 0.8695652 1
## [1] "Sensitivity"
## .id variable ACC BLCA BRCA CESC COLORECTAL DLBC
## 1 rda Sensitivity 0.9565217 0.8196721 0.9755352 0.7802198 1 0.995
## GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER PCPG PRAD
## 1 0.9906191 0.8618421 0.9615385 0.9444444 0.9428571 1 0.9778761 1 0.974026
## SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## 1 0.9666667 0.8715084 0.8589744 0.95 1 0.9444444 0.8695652 1
## [1] "Specificity"
## .id variable ACC BLCA BRCA CESC COLORECTAL DLBC
## 2 rda Specificity 0.9996943 0.9962169 0.999663 0.9953169 0.9996951 1
## GLIOMA HNSC KIDNEY LUNG MESO OV PARALIVER PCPG
## 2 0.9996378 0.9886288 1 0.9981273 0.9993588 0.9996915 0.9967405 1
## PRAD SARC SKCM STOPH TGCT THCA THYM UTERINE UVM
## 2 0.9934722 0.9996936 0.9961477 0.9847036 1 1 1 0.9953488 0.9996942
ctrl = trainControl(method = "repeatedcv", # repeated K-folds
number = 10, # 10 folds
repeats = 10, # 10 repeats
summaryFunction = multiClassSummary, # Evaluate Performance
classProbs = T, # Estimate class probabilities
savePredictions = T,
verboseIter = T)rdaGrid = data.frame(gamma = (1:4)/4, lambda = 0.75)
set.seed(1234)
rdafit = train(mod_id ~ ., data = pptraining, method = "rda", tuneGrid = rdaGrid, trControl = ctrl)## + Fold01.Rep01: gamma=0.25, lambda=0.75
## - Fold01.Rep01: gamma=0.25, lambda=0.75
## + Fold01.Rep01: gamma=0.50, lambda=0.75
## - Fold01.Rep01: gamma=0.50, lambda=0.75
## + Fold01.Rep01: gamma=0.75, lambda=0.75
## - Fold01.Rep01: gamma=0.75, lambda=0.75
## + Fold01.Rep01: gamma=1.00, lambda=0.75
## - Fold01.Rep01: gamma=1.00, lambda=0.75
## + Fold02.Rep01: gamma=0.25, lambda=0.75
## - Fold02.Rep01: gamma=0.25, lambda=0.75
## + Fold02.Rep01: gamma=0.50, lambda=0.75
## - Fold02.Rep01: gamma=0.50, lambda=0.75
## + Fold02.Rep01: gamma=0.75, lambda=0.75
## - Fold02.Rep01: gamma=0.75, lambda=0.75
## + Fold02.Rep01: gamma=1.00, lambda=0.75
## - Fold02.Rep01: gamma=1.00, lambda=0.75
## + Fold03.Rep01: gamma=0.25, lambda=0.75
## - Fold03.Rep01: gamma=0.25, lambda=0.75
## + Fold03.Rep01: gamma=0.50, lambda=0.75
## - Fold03.Rep01: gamma=0.50, lambda=0.75
## + Fold03.Rep01: gamma=0.75, lambda=0.75
## - Fold03.Rep01: gamma=0.75, lambda=0.75
## + Fold03.Rep01: gamma=1.00, lambda=0.75
## - Fold03.Rep01: gamma=1.00, lambda=0.75
## + Fold04.Rep01: gamma=0.25, lambda=0.75
## - Fold04.Rep01: gamma=0.25, lambda=0.75
## + Fold04.Rep01: gamma=0.50, lambda=0.75
## - Fold04.Rep01: gamma=0.50, lambda=0.75
## + Fold04.Rep01: gamma=0.75, lambda=0.75
## - Fold04.Rep01: gamma=0.75, lambda=0.75
## + Fold04.Rep01: gamma=1.00, lambda=0.75
## - Fold04.Rep01: gamma=1.00, lambda=0.75
## + Fold05.Rep01: gamma=0.25, lambda=0.75
## - Fold05.Rep01: gamma=0.25, lambda=0.75
## + Fold05.Rep01: gamma=0.50, lambda=0.75
## - Fold05.Rep01: gamma=0.50, lambda=0.75
## + Fold05.Rep01: gamma=0.75, lambda=0.75
## - Fold05.Rep01: gamma=0.75, lambda=0.75
## + Fold05.Rep01: gamma=1.00, lambda=0.75
## - Fold05.Rep01: gamma=1.00, lambda=0.75
## + Fold06.Rep01: gamma=0.25, lambda=0.75
## - Fold06.Rep01: gamma=0.25, lambda=0.75
## + Fold06.Rep01: gamma=0.50, lambda=0.75
## - Fold06.Rep01: gamma=0.50, lambda=0.75
## + Fold06.Rep01: gamma=0.75, lambda=0.75
## - Fold06.Rep01: gamma=0.75, lambda=0.75
## + Fold06.Rep01: gamma=1.00, lambda=0.75
## - Fold06.Rep01: gamma=1.00, lambda=0.75
## + Fold07.Rep01: gamma=0.25, lambda=0.75
## - Fold07.Rep01: gamma=0.25, lambda=0.75
## + Fold07.Rep01: gamma=0.50, lambda=0.75
## - Fold07.Rep01: gamma=0.50, lambda=0.75
## + Fold07.Rep01: gamma=0.75, lambda=0.75
## - Fold07.Rep01: gamma=0.75, lambda=0.75
## + Fold07.Rep01: gamma=1.00, lambda=0.75
## - Fold07.Rep01: gamma=1.00, lambda=0.75
## + Fold08.Rep01: gamma=0.25, lambda=0.75
## - Fold08.Rep01: gamma=0.25, lambda=0.75
## + Fold08.Rep01: gamma=0.50, lambda=0.75
## - Fold08.Rep01: gamma=0.50, lambda=0.75
## + Fold08.Rep01: gamma=0.75, lambda=0.75
## - Fold08.Rep01: gamma=0.75, lambda=0.75
## + Fold08.Rep01: gamma=1.00, lambda=0.75
## - Fold08.Rep01: gamma=1.00, lambda=0.75
## + Fold09.Rep01: gamma=0.25, lambda=0.75
## - Fold09.Rep01: gamma=0.25, lambda=0.75
## + Fold09.Rep01: gamma=0.50, lambda=0.75
## - Fold09.Rep01: gamma=0.50, lambda=0.75
## + Fold09.Rep01: gamma=0.75, lambda=0.75
## - Fold09.Rep01: gamma=0.75, lambda=0.75
## + Fold09.Rep01: gamma=1.00, lambda=0.75
## - Fold09.Rep01: gamma=1.00, lambda=0.75
## + Fold10.Rep01: gamma=0.25, lambda=0.75
## - Fold10.Rep01: gamma=0.25, lambda=0.75
## + Fold10.Rep01: gamma=0.50, lambda=0.75
## - Fold10.Rep01: gamma=0.50, lambda=0.75
## + Fold10.Rep01: gamma=0.75, lambda=0.75
## - Fold10.Rep01: gamma=0.75, lambda=0.75
## + Fold10.Rep01: gamma=1.00, lambda=0.75
## - Fold10.Rep01: gamma=1.00, lambda=0.75
## + Fold01.Rep02: gamma=0.25, lambda=0.75
## - Fold01.Rep02: gamma=0.25, lambda=0.75
## + Fold01.Rep02: gamma=0.50, lambda=0.75
## - Fold01.Rep02: gamma=0.50, lambda=0.75
## + Fold01.Rep02: gamma=0.75, lambda=0.75
## - Fold01.Rep02: gamma=0.75, lambda=0.75
## + Fold01.Rep02: gamma=1.00, lambda=0.75
## - Fold01.Rep02: gamma=1.00, lambda=0.75
## + Fold02.Rep02: gamma=0.25, lambda=0.75
## - Fold02.Rep02: gamma=0.25, lambda=0.75
## + Fold02.Rep02: gamma=0.50, lambda=0.75
## - Fold02.Rep02: gamma=0.50, lambda=0.75
## + Fold02.Rep02: gamma=0.75, lambda=0.75
## - Fold02.Rep02: gamma=0.75, lambda=0.75
## + Fold02.Rep02: gamma=1.00, lambda=0.75
## - Fold02.Rep02: gamma=1.00, lambda=0.75
## + Fold03.Rep02: gamma=0.25, lambda=0.75
## - Fold03.Rep02: gamma=0.25, lambda=0.75
## + Fold03.Rep02: gamma=0.50, lambda=0.75
## - Fold03.Rep02: gamma=0.50, lambda=0.75
## + Fold03.Rep02: gamma=0.75, lambda=0.75
## - Fold03.Rep02: gamma=0.75, lambda=0.75
## + Fold03.Rep02: gamma=1.00, lambda=0.75
## - Fold03.Rep02: gamma=1.00, lambda=0.75
## + Fold04.Rep02: gamma=0.25, lambda=0.75
## - Fold04.Rep02: gamma=0.25, lambda=0.75
## + Fold04.Rep02: gamma=0.50, lambda=0.75
## - Fold04.Rep02: gamma=0.50, lambda=0.75
## + Fold04.Rep02: gamma=0.75, lambda=0.75
## - Fold04.Rep02: gamma=0.75, lambda=0.75
## + Fold04.Rep02: gamma=1.00, lambda=0.75
## - Fold04.Rep02: gamma=1.00, lambda=0.75
## + Fold05.Rep02: gamma=0.25, lambda=0.75
## - Fold05.Rep02: gamma=0.25, lambda=0.75
## + Fold05.Rep02: gamma=0.50, lambda=0.75
## - Fold05.Rep02: gamma=0.50, lambda=0.75
## + Fold05.Rep02: gamma=0.75, lambda=0.75
## - Fold05.Rep02: gamma=0.75, lambda=0.75
## + Fold05.Rep02: gamma=1.00, lambda=0.75
## - Fold05.Rep02: gamma=1.00, lambda=0.75
## + Fold06.Rep02: gamma=0.25, lambda=0.75
## - Fold06.Rep02: gamma=0.25, lambda=0.75
## + Fold06.Rep02: gamma=0.50, lambda=0.75
## - Fold06.Rep02: gamma=0.50, lambda=0.75
## + Fold06.Rep02: gamma=0.75, lambda=0.75
## - Fold06.Rep02: gamma=0.75, lambda=0.75
## + Fold06.Rep02: gamma=1.00, lambda=0.75
## - Fold06.Rep02: gamma=1.00, lambda=0.75
## + Fold07.Rep02: gamma=0.25, lambda=0.75
## - Fold07.Rep02: gamma=0.25, lambda=0.75
## + Fold07.Rep02: gamma=0.50, lambda=0.75
## - Fold07.Rep02: gamma=0.50, lambda=0.75
## + Fold07.Rep02: gamma=0.75, lambda=0.75
## - Fold07.Rep02: gamma=0.75, lambda=0.75
## + Fold07.Rep02: gamma=1.00, lambda=0.75
## - Fold07.Rep02: gamma=1.00, lambda=0.75
## + Fold08.Rep02: gamma=0.25, lambda=0.75
## - Fold08.Rep02: gamma=0.25, lambda=0.75
## + Fold08.Rep02: gamma=0.50, lambda=0.75
## - Fold08.Rep02: gamma=0.50, lambda=0.75
## + Fold08.Rep02: gamma=0.75, lambda=0.75
## - Fold08.Rep02: gamma=0.75, lambda=0.75
## + Fold08.Rep02: gamma=1.00, lambda=0.75
## - Fold08.Rep02: gamma=1.00, lambda=0.75
## + Fold09.Rep02: gamma=0.25, lambda=0.75
## - Fold09.Rep02: gamma=0.25, lambda=0.75
## + Fold09.Rep02: gamma=0.50, lambda=0.75
## - Fold09.Rep02: gamma=0.50, lambda=0.75
## + Fold09.Rep02: gamma=0.75, lambda=0.75
## - Fold09.Rep02: gamma=0.75, lambda=0.75
## + Fold09.Rep02: gamma=1.00, lambda=0.75
## - Fold09.Rep02: gamma=1.00, lambda=0.75
## + Fold10.Rep02: gamma=0.25, lambda=0.75
## - Fold10.Rep02: gamma=0.25, lambda=0.75
## + Fold10.Rep02: gamma=0.50, lambda=0.75
## - Fold10.Rep02: gamma=0.50, lambda=0.75
## + Fold10.Rep02: gamma=0.75, lambda=0.75
## - Fold10.Rep02: gamma=0.75, lambda=0.75
## + Fold10.Rep02: gamma=1.00, lambda=0.75
## - Fold10.Rep02: gamma=1.00, lambda=0.75
## + Fold01.Rep03: gamma=0.25, lambda=0.75
## - Fold01.Rep03: gamma=0.25, lambda=0.75
## + Fold01.Rep03: gamma=0.50, lambda=0.75
## - Fold01.Rep03: gamma=0.50, lambda=0.75
## + Fold01.Rep03: gamma=0.75, lambda=0.75
## - Fold01.Rep03: gamma=0.75, lambda=0.75
## + Fold01.Rep03: gamma=1.00, lambda=0.75
## - Fold01.Rep03: gamma=1.00, lambda=0.75
## + Fold02.Rep03: gamma=0.25, lambda=0.75
## - Fold02.Rep03: gamma=0.25, lambda=0.75
## + Fold02.Rep03: gamma=0.50, lambda=0.75
## - Fold02.Rep03: gamma=0.50, lambda=0.75
## + Fold02.Rep03: gamma=0.75, lambda=0.75
## - Fold02.Rep03: gamma=0.75, lambda=0.75
## + Fold02.Rep03: gamma=1.00, lambda=0.75
## - Fold02.Rep03: gamma=1.00, lambda=0.75
## + Fold03.Rep03: gamma=0.25, lambda=0.75
## - Fold03.Rep03: gamma=0.25, lambda=0.75
## + Fold03.Rep03: gamma=0.50, lambda=0.75
## - Fold03.Rep03: gamma=0.50, lambda=0.75
## + Fold03.Rep03: gamma=0.75, lambda=0.75
## - Fold03.Rep03: gamma=0.75, lambda=0.75
## + Fold03.Rep03: gamma=1.00, lambda=0.75
## - Fold03.Rep03: gamma=1.00, lambda=0.75
## + Fold04.Rep03: gamma=0.25, lambda=0.75
## - Fold04.Rep03: gamma=0.25, lambda=0.75
## + Fold04.Rep03: gamma=0.50, lambda=0.75
## - Fold04.Rep03: gamma=0.50, lambda=0.75
## + Fold04.Rep03: gamma=0.75, lambda=0.75
## - Fold04.Rep03: gamma=0.75, lambda=0.75
## + Fold04.Rep03: gamma=1.00, lambda=0.75
## - Fold04.Rep03: gamma=1.00, lambda=0.75
## + Fold05.Rep03: gamma=0.25, lambda=0.75
## - Fold05.Rep03: gamma=0.25, lambda=0.75
## + Fold05.Rep03: gamma=0.50, lambda=0.75
## - Fold05.Rep03: gamma=0.50, lambda=0.75
## + Fold05.Rep03: gamma=0.75, lambda=0.75
## - Fold05.Rep03: gamma=0.75, lambda=0.75
## + Fold05.Rep03: gamma=1.00, lambda=0.75
## - Fold05.Rep03: gamma=1.00, lambda=0.75
## + Fold06.Rep03: gamma=0.25, lambda=0.75
## - Fold06.Rep03: gamma=0.25, lambda=0.75
## + Fold06.Rep03: gamma=0.50, lambda=0.75
## - Fold06.Rep03: gamma=0.50, lambda=0.75
## + Fold06.Rep03: gamma=0.75, lambda=0.75
## - Fold06.Rep03: gamma=0.75, lambda=0.75
## + Fold06.Rep03: gamma=1.00, lambda=0.75
## - Fold06.Rep03: gamma=1.00, lambda=0.75
## + Fold07.Rep03: gamma=0.25, lambda=0.75
## - Fold07.Rep03: gamma=0.25, lambda=0.75
## + Fold07.Rep03: gamma=0.50, lambda=0.75
## - Fold07.Rep03: gamma=0.50, lambda=0.75
## + Fold07.Rep03: gamma=0.75, lambda=0.75
## - Fold07.Rep03: gamma=0.75, lambda=0.75
## + Fold07.Rep03: gamma=1.00, lambda=0.75
## - Fold07.Rep03: gamma=1.00, lambda=0.75
## + Fold08.Rep03: gamma=0.25, lambda=0.75
## - Fold08.Rep03: gamma=0.25, lambda=0.75
## + Fold08.Rep03: gamma=0.50, lambda=0.75
## - Fold08.Rep03: gamma=0.50, lambda=0.75
## + Fold08.Rep03: gamma=0.75, lambda=0.75
## - Fold08.Rep03: gamma=0.75, lambda=0.75
## + Fold08.Rep03: gamma=1.00, lambda=0.75
## - Fold08.Rep03: gamma=1.00, lambda=0.75
## + Fold09.Rep03: gamma=0.25, lambda=0.75
## - Fold09.Rep03: gamma=0.25, lambda=0.75
## + Fold09.Rep03: gamma=0.50, lambda=0.75
## - Fold09.Rep03: gamma=0.50, lambda=0.75
## + Fold09.Rep03: gamma=0.75, lambda=0.75
## - Fold09.Rep03: gamma=0.75, lambda=0.75
## + Fold09.Rep03: gamma=1.00, lambda=0.75
## - Fold09.Rep03: gamma=1.00, lambda=0.75
## + Fold10.Rep03: gamma=0.25, lambda=0.75
## - Fold10.Rep03: gamma=0.25, lambda=0.75
## + Fold10.Rep03: gamma=0.50, lambda=0.75
## - Fold10.Rep03: gamma=0.50, lambda=0.75
## + Fold10.Rep03: gamma=0.75, lambda=0.75
## - Fold10.Rep03: gamma=0.75, lambda=0.75
## + Fold10.Rep03: gamma=1.00, lambda=0.75
## - Fold10.Rep03: gamma=1.00, lambda=0.75
## + Fold01.Rep04: gamma=0.25, lambda=0.75
## - Fold01.Rep04: gamma=0.25, lambda=0.75
## + Fold01.Rep04: gamma=0.50, lambda=0.75
## - Fold01.Rep04: gamma=0.50, lambda=0.75
## + Fold01.Rep04: gamma=0.75, lambda=0.75
## - Fold01.Rep04: gamma=0.75, lambda=0.75
## + Fold01.Rep04: gamma=1.00, lambda=0.75
## - Fold01.Rep04: gamma=1.00, lambda=0.75
## + Fold02.Rep04: gamma=0.25, lambda=0.75
## - Fold02.Rep04: gamma=0.25, lambda=0.75
## + Fold02.Rep04: gamma=0.50, lambda=0.75
## - Fold02.Rep04: gamma=0.50, lambda=0.75
## + Fold02.Rep04: gamma=0.75, lambda=0.75
## - Fold02.Rep04: gamma=0.75, lambda=0.75
## + Fold02.Rep04: gamma=1.00, lambda=0.75
## - Fold02.Rep04: gamma=1.00, lambda=0.75
## + Fold03.Rep04: gamma=0.25, lambda=0.75
## - Fold03.Rep04: gamma=0.25, lambda=0.75
## + Fold03.Rep04: gamma=0.50, lambda=0.75
## - Fold03.Rep04: gamma=0.50, lambda=0.75
## + Fold03.Rep04: gamma=0.75, lambda=0.75
## - Fold03.Rep04: gamma=0.75, lambda=0.75
## + Fold03.Rep04: gamma=1.00, lambda=0.75
## - Fold03.Rep04: gamma=1.00, lambda=0.75
## + Fold04.Rep04: gamma=0.25, lambda=0.75
## - Fold04.Rep04: gamma=0.25, lambda=0.75
## + Fold04.Rep04: gamma=0.50, lambda=0.75
## - Fold04.Rep04: gamma=0.50, lambda=0.75
## + Fold04.Rep04: gamma=0.75, lambda=0.75
## - Fold04.Rep04: gamma=0.75, lambda=0.75
## + Fold04.Rep04: gamma=1.00, lambda=0.75
## - Fold04.Rep04: gamma=1.00, lambda=0.75
## + Fold05.Rep04: gamma=0.25, lambda=0.75
## - Fold05.Rep04: gamma=0.25, lambda=0.75
## + Fold05.Rep04: gamma=0.50, lambda=0.75
## - Fold05.Rep04: gamma=0.50, lambda=0.75
## + Fold05.Rep04: gamma=0.75, lambda=0.75
## - Fold05.Rep04: gamma=0.75, lambda=0.75
## + Fold05.Rep04: gamma=1.00, lambda=0.75
## - Fold05.Rep04: gamma=1.00, lambda=0.75
## + Fold06.Rep04: gamma=0.25, lambda=0.75
## - Fold06.Rep04: gamma=0.25, lambda=0.75
## + Fold06.Rep04: gamma=0.50, lambda=0.75
## - Fold06.Rep04: gamma=0.50, lambda=0.75
## + Fold06.Rep04: gamma=0.75, lambda=0.75
## - Fold06.Rep04: gamma=0.75, lambda=0.75
## + Fold06.Rep04: gamma=1.00, lambda=0.75
## - Fold06.Rep04: gamma=1.00, lambda=0.75
## + Fold07.Rep04: gamma=0.25, lambda=0.75
## - Fold07.Rep04: gamma=0.25, lambda=0.75
## + Fold07.Rep04: gamma=0.50, lambda=0.75
## - Fold07.Rep04: gamma=0.50, lambda=0.75
## + Fold07.Rep04: gamma=0.75, lambda=0.75
## - Fold07.Rep04: gamma=0.75, lambda=0.75
## + Fold07.Rep04: gamma=1.00, lambda=0.75
## - Fold07.Rep04: gamma=1.00, lambda=0.75
## + Fold08.Rep04: gamma=0.25, lambda=0.75
## - Fold08.Rep04: gamma=0.25, lambda=0.75
## + Fold08.Rep04: gamma=0.50, lambda=0.75
## - Fold08.Rep04: gamma=0.50, lambda=0.75
## + Fold08.Rep04: gamma=0.75, lambda=0.75
## - Fold08.Rep04: gamma=0.75, lambda=0.75
## + Fold08.Rep04: gamma=1.00, lambda=0.75
## - Fold08.Rep04: gamma=1.00, lambda=0.75
## + Fold09.Rep04: gamma=0.25, lambda=0.75
## - Fold09.Rep04: gamma=0.25, lambda=0.75
## + Fold09.Rep04: gamma=0.50, lambda=0.75
## - Fold09.Rep04: gamma=0.50, lambda=0.75
## + Fold09.Rep04: gamma=0.75, lambda=0.75
## - Fold09.Rep04: gamma=0.75, lambda=0.75
## + Fold09.Rep04: gamma=1.00, lambda=0.75
## - Fold09.Rep04: gamma=1.00, lambda=0.75
## + Fold10.Rep04: gamma=0.25, lambda=0.75
## - Fold10.Rep04: gamma=0.25, lambda=0.75
## + Fold10.Rep04: gamma=0.50, lambda=0.75
## - Fold10.Rep04: gamma=0.50, lambda=0.75
## + Fold10.Rep04: gamma=0.75, lambda=0.75
## - Fold10.Rep04: gamma=0.75, lambda=0.75
## + Fold10.Rep04: gamma=1.00, lambda=0.75
## - Fold10.Rep04: gamma=1.00, lambda=0.75
## + Fold01.Rep05: gamma=0.25, lambda=0.75
## - Fold01.Rep05: gamma=0.25, lambda=0.75
## + Fold01.Rep05: gamma=0.50, lambda=0.75
## - Fold01.Rep05: gamma=0.50, lambda=0.75
## + Fold01.Rep05: gamma=0.75, lambda=0.75
## - Fold01.Rep05: gamma=0.75, lambda=0.75
## + Fold01.Rep05: gamma=1.00, lambda=0.75
## - Fold01.Rep05: gamma=1.00, lambda=0.75
## + Fold02.Rep05: gamma=0.25, lambda=0.75
## - Fold02.Rep05: gamma=0.25, lambda=0.75
## + Fold02.Rep05: gamma=0.50, lambda=0.75
## - Fold02.Rep05: gamma=0.50, lambda=0.75
## + Fold02.Rep05: gamma=0.75, lambda=0.75
## - Fold02.Rep05: gamma=0.75, lambda=0.75
## + Fold02.Rep05: gamma=1.00, lambda=0.75
## - Fold02.Rep05: gamma=1.00, lambda=0.75
## + Fold03.Rep05: gamma=0.25, lambda=0.75
## - Fold03.Rep05: gamma=0.25, lambda=0.75
## + Fold03.Rep05: gamma=0.50, lambda=0.75
## - Fold03.Rep05: gamma=0.50, lambda=0.75
## + Fold03.Rep05: gamma=0.75, lambda=0.75
## - Fold03.Rep05: gamma=0.75, lambda=0.75
## + Fold03.Rep05: gamma=1.00, lambda=0.75
## - Fold03.Rep05: gamma=1.00, lambda=0.75
## + Fold04.Rep05: gamma=0.25, lambda=0.75
## - Fold04.Rep05: gamma=0.25, lambda=0.75
## + Fold04.Rep05: gamma=0.50, lambda=0.75
## - Fold04.Rep05: gamma=0.50, lambda=0.75
## + Fold04.Rep05: gamma=0.75, lambda=0.75
## - Fold04.Rep05: gamma=0.75, lambda=0.75
## + Fold04.Rep05: gamma=1.00, lambda=0.75
## - Fold04.Rep05: gamma=1.00, lambda=0.75
## + Fold05.Rep05: gamma=0.25, lambda=0.75
## - Fold05.Rep05: gamma=0.25, lambda=0.75
## + Fold05.Rep05: gamma=0.50, lambda=0.75
## - Fold05.Rep05: gamma=0.50, lambda=0.75
## + Fold05.Rep05: gamma=0.75, lambda=0.75
## - Fold05.Rep05: gamma=0.75, lambda=0.75
## + Fold05.Rep05: gamma=1.00, lambda=0.75
## - Fold05.Rep05: gamma=1.00, lambda=0.75
## + Fold06.Rep05: gamma=0.25, lambda=0.75
## - Fold06.Rep05: gamma=0.25, lambda=0.75
## + Fold06.Rep05: gamma=0.50, lambda=0.75
## - Fold06.Rep05: gamma=0.50, lambda=0.75
## + Fold06.Rep05: gamma=0.75, lambda=0.75
## - Fold06.Rep05: gamma=0.75, lambda=0.75
## + Fold06.Rep05: gamma=1.00, lambda=0.75
## - Fold06.Rep05: gamma=1.00, lambda=0.75
## + Fold07.Rep05: gamma=0.25, lambda=0.75
## - Fold07.Rep05: gamma=0.25, lambda=0.75
## + Fold07.Rep05: gamma=0.50, lambda=0.75
## - Fold07.Rep05: gamma=0.50, lambda=0.75
## + Fold07.Rep05: gamma=0.75, lambda=0.75
## - Fold07.Rep05: gamma=0.75, lambda=0.75
## + Fold07.Rep05: gamma=1.00, lambda=0.75
## - Fold07.Rep05: gamma=1.00, lambda=0.75
## + Fold08.Rep05: gamma=0.25, lambda=0.75
## - Fold08.Rep05: gamma=0.25, lambda=0.75
## + Fold08.Rep05: gamma=0.50, lambda=0.75
## - Fold08.Rep05: gamma=0.50, lambda=0.75
## + Fold08.Rep05: gamma=0.75, lambda=0.75
## - Fold08.Rep05: gamma=0.75, lambda=0.75
## + Fold08.Rep05: gamma=1.00, lambda=0.75
## - Fold08.Rep05: gamma=1.00, lambda=0.75
## + Fold09.Rep05: gamma=0.25, lambda=0.75
## - Fold09.Rep05: gamma=0.25, lambda=0.75
## + Fold09.Rep05: gamma=0.50, lambda=0.75
## - Fold09.Rep05: gamma=0.50, lambda=0.75
## + Fold09.Rep05: gamma=0.75, lambda=0.75
## - Fold09.Rep05: gamma=0.75, lambda=0.75
## + Fold09.Rep05: gamma=1.00, lambda=0.75
## - Fold09.Rep05: gamma=1.00, lambda=0.75
## + Fold10.Rep05: gamma=0.25, lambda=0.75
## - Fold10.Rep05: gamma=0.25, lambda=0.75
## + Fold10.Rep05: gamma=0.50, lambda=0.75
## - Fold10.Rep05: gamma=0.50, lambda=0.75
## + Fold10.Rep05: gamma=0.75, lambda=0.75
## - Fold10.Rep05: gamma=0.75, lambda=0.75
## + Fold10.Rep05: gamma=1.00, lambda=0.75
## - Fold10.Rep05: gamma=1.00, lambda=0.75
## + Fold01.Rep06: gamma=0.25, lambda=0.75
## - Fold01.Rep06: gamma=0.25, lambda=0.75
## + Fold01.Rep06: gamma=0.50, lambda=0.75
## - Fold01.Rep06: gamma=0.50, lambda=0.75
## + Fold01.Rep06: gamma=0.75, lambda=0.75
## - Fold01.Rep06: gamma=0.75, lambda=0.75
## + Fold01.Rep06: gamma=1.00, lambda=0.75
## - Fold01.Rep06: gamma=1.00, lambda=0.75
## + Fold02.Rep06: gamma=0.25, lambda=0.75
## - Fold02.Rep06: gamma=0.25, lambda=0.75
## + Fold02.Rep06: gamma=0.50, lambda=0.75
## - Fold02.Rep06: gamma=0.50, lambda=0.75
## + Fold02.Rep06: gamma=0.75, lambda=0.75
## - Fold02.Rep06: gamma=0.75, lambda=0.75
## + Fold02.Rep06: gamma=1.00, lambda=0.75
## - Fold02.Rep06: gamma=1.00, lambda=0.75
## + Fold03.Rep06: gamma=0.25, lambda=0.75
## - Fold03.Rep06: gamma=0.25, lambda=0.75
## + Fold03.Rep06: gamma=0.50, lambda=0.75
## - Fold03.Rep06: gamma=0.50, lambda=0.75
## + Fold03.Rep06: gamma=0.75, lambda=0.75
## - Fold03.Rep06: gamma=0.75, lambda=0.75
## + Fold03.Rep06: gamma=1.00, lambda=0.75
## - Fold03.Rep06: gamma=1.00, lambda=0.75
## + Fold04.Rep06: gamma=0.25, lambda=0.75
## - Fold04.Rep06: gamma=0.25, lambda=0.75
## + Fold04.Rep06: gamma=0.50, lambda=0.75
## - Fold04.Rep06: gamma=0.50, lambda=0.75
## + Fold04.Rep06: gamma=0.75, lambda=0.75
## - Fold04.Rep06: gamma=0.75, lambda=0.75
## + Fold04.Rep06: gamma=1.00, lambda=0.75
## - Fold04.Rep06: gamma=1.00, lambda=0.75
## + Fold05.Rep06: gamma=0.25, lambda=0.75
## - Fold05.Rep06: gamma=0.25, lambda=0.75
## + Fold05.Rep06: gamma=0.50, lambda=0.75
## - Fold05.Rep06: gamma=0.50, lambda=0.75
## + Fold05.Rep06: gamma=0.75, lambda=0.75
## - Fold05.Rep06: gamma=0.75, lambda=0.75
## + Fold05.Rep06: gamma=1.00, lambda=0.75
## - Fold05.Rep06: gamma=1.00, lambda=0.75
## + Fold06.Rep06: gamma=0.25, lambda=0.75
## - Fold06.Rep06: gamma=0.25, lambda=0.75
## + Fold06.Rep06: gamma=0.50, lambda=0.75
## - Fold06.Rep06: gamma=0.50, lambda=0.75
## + Fold06.Rep06: gamma=0.75, lambda=0.75
## - Fold06.Rep06: gamma=0.75, lambda=0.75
## + Fold06.Rep06: gamma=1.00, lambda=0.75
## - Fold06.Rep06: gamma=1.00, lambda=0.75
## + Fold07.Rep06: gamma=0.25, lambda=0.75
## - Fold07.Rep06: gamma=0.25, lambda=0.75
## + Fold07.Rep06: gamma=0.50, lambda=0.75
## - Fold07.Rep06: gamma=0.50, lambda=0.75
## + Fold07.Rep06: gamma=0.75, lambda=0.75
## - Fold07.Rep06: gamma=0.75, lambda=0.75
## + Fold07.Rep06: gamma=1.00, lambda=0.75
## - Fold07.Rep06: gamma=1.00, lambda=0.75
## + Fold08.Rep06: gamma=0.25, lambda=0.75
## - Fold08.Rep06: gamma=0.25, lambda=0.75
## + Fold08.Rep06: gamma=0.50, lambda=0.75
## - Fold08.Rep06: gamma=0.50, lambda=0.75
## + Fold08.Rep06: gamma=0.75, lambda=0.75
## - Fold08.Rep06: gamma=0.75, lambda=0.75
## + Fold08.Rep06: gamma=1.00, lambda=0.75
## - Fold08.Rep06: gamma=1.00, lambda=0.75
## + Fold09.Rep06: gamma=0.25, lambda=0.75
## - Fold09.Rep06: gamma=0.25, lambda=0.75
## + Fold09.Rep06: gamma=0.50, lambda=0.75
## - Fold09.Rep06: gamma=0.50, lambda=0.75
## + Fold09.Rep06: gamma=0.75, lambda=0.75
## - Fold09.Rep06: gamma=0.75, lambda=0.75
## + Fold09.Rep06: gamma=1.00, lambda=0.75
## - Fold09.Rep06: gamma=1.00, lambda=0.75
## + Fold10.Rep06: gamma=0.25, lambda=0.75
## - Fold10.Rep06: gamma=0.25, lambda=0.75
## + Fold10.Rep06: gamma=0.50, lambda=0.75
## - Fold10.Rep06: gamma=0.50, lambda=0.75
## + Fold10.Rep06: gamma=0.75, lambda=0.75
## - Fold10.Rep06: gamma=0.75, lambda=0.75
## + Fold10.Rep06: gamma=1.00, lambda=0.75
## - Fold10.Rep06: gamma=1.00, lambda=0.75
## + Fold01.Rep07: gamma=0.25, lambda=0.75
## - Fold01.Rep07: gamma=0.25, lambda=0.75
## + Fold01.Rep07: gamma=0.50, lambda=0.75
## - Fold01.Rep07: gamma=0.50, lambda=0.75
## + Fold01.Rep07: gamma=0.75, lambda=0.75
## - Fold01.Rep07: gamma=0.75, lambda=0.75
## + Fold01.Rep07: gamma=1.00, lambda=0.75
## - Fold01.Rep07: gamma=1.00, lambda=0.75
## + Fold02.Rep07: gamma=0.25, lambda=0.75
## - Fold02.Rep07: gamma=0.25, lambda=0.75
## + Fold02.Rep07: gamma=0.50, lambda=0.75
## - Fold02.Rep07: gamma=0.50, lambda=0.75
## + Fold02.Rep07: gamma=0.75, lambda=0.75
## - Fold02.Rep07: gamma=0.75, lambda=0.75
## + Fold02.Rep07: gamma=1.00, lambda=0.75
## - Fold02.Rep07: gamma=1.00, lambda=0.75
## + Fold03.Rep07: gamma=0.25, lambda=0.75
## - Fold03.Rep07: gamma=0.25, lambda=0.75
## + Fold03.Rep07: gamma=0.50, lambda=0.75
## - Fold03.Rep07: gamma=0.50, lambda=0.75
## + Fold03.Rep07: gamma=0.75, lambda=0.75
## - Fold03.Rep07: gamma=0.75, lambda=0.75
## + Fold03.Rep07: gamma=1.00, lambda=0.75
## - Fold03.Rep07: gamma=1.00, lambda=0.75
## + Fold04.Rep07: gamma=0.25, lambda=0.75
## - Fold04.Rep07: gamma=0.25, lambda=0.75
## + Fold04.Rep07: gamma=0.50, lambda=0.75
## - Fold04.Rep07: gamma=0.50, lambda=0.75
## + Fold04.Rep07: gamma=0.75, lambda=0.75
## - Fold04.Rep07: gamma=0.75, lambda=0.75
## + Fold04.Rep07: gamma=1.00, lambda=0.75
## - Fold04.Rep07: gamma=1.00, lambda=0.75
## + Fold05.Rep07: gamma=0.25, lambda=0.75
## - Fold05.Rep07: gamma=0.25, lambda=0.75
## + Fold05.Rep07: gamma=0.50, lambda=0.75
## - Fold05.Rep07: gamma=0.50, lambda=0.75
## + Fold05.Rep07: gamma=0.75, lambda=0.75
## - Fold05.Rep07: gamma=0.75, lambda=0.75
## + Fold05.Rep07: gamma=1.00, lambda=0.75
## - Fold05.Rep07: gamma=1.00, lambda=0.75
## + Fold06.Rep07: gamma=0.25, lambda=0.75
## - Fold06.Rep07: gamma=0.25, lambda=0.75
## + Fold06.Rep07: gamma=0.50, lambda=0.75
## - Fold06.Rep07: gamma=0.50, lambda=0.75
## + Fold06.Rep07: gamma=0.75, lambda=0.75
## - Fold06.Rep07: gamma=0.75, lambda=0.75
## + Fold06.Rep07: gamma=1.00, lambda=0.75
## - Fold06.Rep07: gamma=1.00, lambda=0.75
## + Fold07.Rep07: gamma=0.25, lambda=0.75
## - Fold07.Rep07: gamma=0.25, lambda=0.75
## + Fold07.Rep07: gamma=0.50, lambda=0.75
## - Fold07.Rep07: gamma=0.50, lambda=0.75
## + Fold07.Rep07: gamma=0.75, lambda=0.75
## - Fold07.Rep07: gamma=0.75, lambda=0.75
## + Fold07.Rep07: gamma=1.00, lambda=0.75
## - Fold07.Rep07: gamma=1.00, lambda=0.75
## + Fold08.Rep07: gamma=0.25, lambda=0.75
## - Fold08.Rep07: gamma=0.25, lambda=0.75
## + Fold08.Rep07: gamma=0.50, lambda=0.75
## - Fold08.Rep07: gamma=0.50, lambda=0.75
## + Fold08.Rep07: gamma=0.75, lambda=0.75
## - Fold08.Rep07: gamma=0.75, lambda=0.75
## + Fold08.Rep07: gamma=1.00, lambda=0.75
## - Fold08.Rep07: gamma=1.00, lambda=0.75
## + Fold09.Rep07: gamma=0.25, lambda=0.75
## - Fold09.Rep07: gamma=0.25, lambda=0.75
## + Fold09.Rep07: gamma=0.50, lambda=0.75
## - Fold09.Rep07: gamma=0.50, lambda=0.75
## + Fold09.Rep07: gamma=0.75, lambda=0.75
## - Fold09.Rep07: gamma=0.75, lambda=0.75
## + Fold09.Rep07: gamma=1.00, lambda=0.75
## - Fold09.Rep07: gamma=1.00, lambda=0.75
## + Fold10.Rep07: gamma=0.25, lambda=0.75
## - Fold10.Rep07: gamma=0.25, lambda=0.75
## + Fold10.Rep07: gamma=0.50, lambda=0.75
## - Fold10.Rep07: gamma=0.50, lambda=0.75
## + Fold10.Rep07: gamma=0.75, lambda=0.75
## - Fold10.Rep07: gamma=0.75, lambda=0.75
## + Fold10.Rep07: gamma=1.00, lambda=0.75
## - Fold10.Rep07: gamma=1.00, lambda=0.75
## + Fold01.Rep08: gamma=0.25, lambda=0.75
## - Fold01.Rep08: gamma=0.25, lambda=0.75
## + Fold01.Rep08: gamma=0.50, lambda=0.75
## - Fold01.Rep08: gamma=0.50, lambda=0.75
## + Fold01.Rep08: gamma=0.75, lambda=0.75
## - Fold01.Rep08: gamma=0.75, lambda=0.75
## + Fold01.Rep08: gamma=1.00, lambda=0.75
## - Fold01.Rep08: gamma=1.00, lambda=0.75
## + Fold02.Rep08: gamma=0.25, lambda=0.75
## - Fold02.Rep08: gamma=0.25, lambda=0.75
## + Fold02.Rep08: gamma=0.50, lambda=0.75
## - Fold02.Rep08: gamma=0.50, lambda=0.75
## + Fold02.Rep08: gamma=0.75, lambda=0.75
## - Fold02.Rep08: gamma=0.75, lambda=0.75
## + Fold02.Rep08: gamma=1.00, lambda=0.75
## - Fold02.Rep08: gamma=1.00, lambda=0.75
## + Fold03.Rep08: gamma=0.25, lambda=0.75
## - Fold03.Rep08: gamma=0.25, lambda=0.75
## + Fold03.Rep08: gamma=0.50, lambda=0.75
## - Fold03.Rep08: gamma=0.50, lambda=0.75
## + Fold03.Rep08: gamma=0.75, lambda=0.75
## - Fold03.Rep08: gamma=0.75, lambda=0.75
## + Fold03.Rep08: gamma=1.00, lambda=0.75
## - Fold03.Rep08: gamma=1.00, lambda=0.75
## + Fold04.Rep08: gamma=0.25, lambda=0.75
## - Fold04.Rep08: gamma=0.25, lambda=0.75
## + Fold04.Rep08: gamma=0.50, lambda=0.75
## - Fold04.Rep08: gamma=0.50, lambda=0.75
## + Fold04.Rep08: gamma=0.75, lambda=0.75
## - Fold04.Rep08: gamma=0.75, lambda=0.75
## + Fold04.Rep08: gamma=1.00, lambda=0.75
## - Fold04.Rep08: gamma=1.00, lambda=0.75
## + Fold05.Rep08: gamma=0.25, lambda=0.75
## - Fold05.Rep08: gamma=0.25, lambda=0.75
## + Fold05.Rep08: gamma=0.50, lambda=0.75
## - Fold05.Rep08: gamma=0.50, lambda=0.75
## + Fold05.Rep08: gamma=0.75, lambda=0.75
## - Fold05.Rep08: gamma=0.75, lambda=0.75
## + Fold05.Rep08: gamma=1.00, lambda=0.75
## - Fold05.Rep08: gamma=1.00, lambda=0.75
## + Fold06.Rep08: gamma=0.25, lambda=0.75
## - Fold06.Rep08: gamma=0.25, lambda=0.75
## + Fold06.Rep08: gamma=0.50, lambda=0.75
## - Fold06.Rep08: gamma=0.50, lambda=0.75
## + Fold06.Rep08: gamma=0.75, lambda=0.75
## - Fold06.Rep08: gamma=0.75, lambda=0.75
## + Fold06.Rep08: gamma=1.00, lambda=0.75
## - Fold06.Rep08: gamma=1.00, lambda=0.75
## + Fold07.Rep08: gamma=0.25, lambda=0.75
## - Fold07.Rep08: gamma=0.25, lambda=0.75
## + Fold07.Rep08: gamma=0.50, lambda=0.75
## - Fold07.Rep08: gamma=0.50, lambda=0.75
## + Fold07.Rep08: gamma=0.75, lambda=0.75
## - Fold07.Rep08: gamma=0.75, lambda=0.75
## + Fold07.Rep08: gamma=1.00, lambda=0.75
## - Fold07.Rep08: gamma=1.00, lambda=0.75
## + Fold08.Rep08: gamma=0.25, lambda=0.75
## - Fold08.Rep08: gamma=0.25, lambda=0.75
## + Fold08.Rep08: gamma=0.50, lambda=0.75
## - Fold08.Rep08: gamma=0.50, lambda=0.75
## + Fold08.Rep08: gamma=0.75, lambda=0.75
## - Fold08.Rep08: gamma=0.75, lambda=0.75
## + Fold08.Rep08: gamma=1.00, lambda=0.75
## - Fold08.Rep08: gamma=1.00, lambda=0.75
## + Fold09.Rep08: gamma=0.25, lambda=0.75
## - Fold09.Rep08: gamma=0.25, lambda=0.75
## + Fold09.Rep08: gamma=0.50, lambda=0.75
## - Fold09.Rep08: gamma=0.50, lambda=0.75
## + Fold09.Rep08: gamma=0.75, lambda=0.75
## - Fold09.Rep08: gamma=0.75, lambda=0.75
## + Fold09.Rep08: gamma=1.00, lambda=0.75
## - Fold09.Rep08: gamma=1.00, lambda=0.75
## + Fold10.Rep08: gamma=0.25, lambda=0.75
## - Fold10.Rep08: gamma=0.25, lambda=0.75
## + Fold10.Rep08: gamma=0.50, lambda=0.75
## - Fold10.Rep08: gamma=0.50, lambda=0.75
## + Fold10.Rep08: gamma=0.75, lambda=0.75
## - Fold10.Rep08: gamma=0.75, lambda=0.75
## + Fold10.Rep08: gamma=1.00, lambda=0.75
## - Fold10.Rep08: gamma=1.00, lambda=0.75
## + Fold01.Rep09: gamma=0.25, lambda=0.75
## - Fold01.Rep09: gamma=0.25, lambda=0.75
## + Fold01.Rep09: gamma=0.50, lambda=0.75
## - Fold01.Rep09: gamma=0.50, lambda=0.75
## + Fold01.Rep09: gamma=0.75, lambda=0.75
## - Fold01.Rep09: gamma=0.75, lambda=0.75
## + Fold01.Rep09: gamma=1.00, lambda=0.75
## - Fold01.Rep09: gamma=1.00, lambda=0.75
## + Fold02.Rep09: gamma=0.25, lambda=0.75
## - Fold02.Rep09: gamma=0.25, lambda=0.75
## + Fold02.Rep09: gamma=0.50, lambda=0.75
## - Fold02.Rep09: gamma=0.50, lambda=0.75
## + Fold02.Rep09: gamma=0.75, lambda=0.75
## - Fold02.Rep09: gamma=0.75, lambda=0.75
## + Fold02.Rep09: gamma=1.00, lambda=0.75
## - Fold02.Rep09: gamma=1.00, lambda=0.75
## + Fold03.Rep09: gamma=0.25, lambda=0.75
## - Fold03.Rep09: gamma=0.25, lambda=0.75
## + Fold03.Rep09: gamma=0.50, lambda=0.75
## - Fold03.Rep09: gamma=0.50, lambda=0.75
## + Fold03.Rep09: gamma=0.75, lambda=0.75
## - Fold03.Rep09: gamma=0.75, lambda=0.75
## + Fold03.Rep09: gamma=1.00, lambda=0.75
## - Fold03.Rep09: gamma=1.00, lambda=0.75
## + Fold04.Rep09: gamma=0.25, lambda=0.75
## - Fold04.Rep09: gamma=0.25, lambda=0.75
## + Fold04.Rep09: gamma=0.50, lambda=0.75
## - Fold04.Rep09: gamma=0.50, lambda=0.75
## + Fold04.Rep09: gamma=0.75, lambda=0.75
## - Fold04.Rep09: gamma=0.75, lambda=0.75
## + Fold04.Rep09: gamma=1.00, lambda=0.75
## - Fold04.Rep09: gamma=1.00, lambda=0.75
## + Fold05.Rep09: gamma=0.25, lambda=0.75
## - Fold05.Rep09: gamma=0.25, lambda=0.75
## + Fold05.Rep09: gamma=0.50, lambda=0.75
## - Fold05.Rep09: gamma=0.50, lambda=0.75
## + Fold05.Rep09: gamma=0.75, lambda=0.75
## - Fold05.Rep09: gamma=0.75, lambda=0.75
## + Fold05.Rep09: gamma=1.00, lambda=0.75
## - Fold05.Rep09: gamma=1.00, lambda=0.75
## + Fold06.Rep09: gamma=0.25, lambda=0.75
## - Fold06.Rep09: gamma=0.25, lambda=0.75
## + Fold06.Rep09: gamma=0.50, lambda=0.75
## - Fold06.Rep09: gamma=0.50, lambda=0.75
## + Fold06.Rep09: gamma=0.75, lambda=0.75
## - Fold06.Rep09: gamma=0.75, lambda=0.75
## + Fold06.Rep09: gamma=1.00, lambda=0.75
## - Fold06.Rep09: gamma=1.00, lambda=0.75
## + Fold07.Rep09: gamma=0.25, lambda=0.75
## - Fold07.Rep09: gamma=0.25, lambda=0.75
## + Fold07.Rep09: gamma=0.50, lambda=0.75
## - Fold07.Rep09: gamma=0.50, lambda=0.75
## + Fold07.Rep09: gamma=0.75, lambda=0.75
## - Fold07.Rep09: gamma=0.75, lambda=0.75
## + Fold07.Rep09: gamma=1.00, lambda=0.75
## - Fold07.Rep09: gamma=1.00, lambda=0.75
## + Fold08.Rep09: gamma=0.25, lambda=0.75
## - Fold08.Rep09: gamma=0.25, lambda=0.75
## + Fold08.Rep09: gamma=0.50, lambda=0.75
## - Fold08.Rep09: gamma=0.50, lambda=0.75
## + Fold08.Rep09: gamma=0.75, lambda=0.75
## - Fold08.Rep09: gamma=0.75, lambda=0.75
## + Fold08.Rep09: gamma=1.00, lambda=0.75
## - Fold08.Rep09: gamma=1.00, lambda=0.75
## + Fold09.Rep09: gamma=0.25, lambda=0.75
## - Fold09.Rep09: gamma=0.25, lambda=0.75
## + Fold09.Rep09: gamma=0.50, lambda=0.75
## - Fold09.Rep09: gamma=0.50, lambda=0.75
## + Fold09.Rep09: gamma=0.75, lambda=0.75
## - Fold09.Rep09: gamma=0.75, lambda=0.75
## + Fold09.Rep09: gamma=1.00, lambda=0.75
## - Fold09.Rep09: gamma=1.00, lambda=0.75
## + Fold10.Rep09: gamma=0.25, lambda=0.75
## - Fold10.Rep09: gamma=0.25, lambda=0.75
## + Fold10.Rep09: gamma=0.50, lambda=0.75
## - Fold10.Rep09: gamma=0.50, lambda=0.75
## + Fold10.Rep09: gamma=0.75, lambda=0.75
## - Fold10.Rep09: gamma=0.75, lambda=0.75
## + Fold10.Rep09: gamma=1.00, lambda=0.75
## - Fold10.Rep09: gamma=1.00, lambda=0.75
## + Fold01.Rep10: gamma=0.25, lambda=0.75
## - Fold01.Rep10: gamma=0.25, lambda=0.75
## + Fold01.Rep10: gamma=0.50, lambda=0.75
## - Fold01.Rep10: gamma=0.50, lambda=0.75
## + Fold01.Rep10: gamma=0.75, lambda=0.75
## - Fold01.Rep10: gamma=0.75, lambda=0.75
## + Fold01.Rep10: gamma=1.00, lambda=0.75
## - Fold01.Rep10: gamma=1.00, lambda=0.75
## + Fold02.Rep10: gamma=0.25, lambda=0.75
## - Fold02.Rep10: gamma=0.25, lambda=0.75
## + Fold02.Rep10: gamma=0.50, lambda=0.75
## - Fold02.Rep10: gamma=0.50, lambda=0.75
## + Fold02.Rep10: gamma=0.75, lambda=0.75
## - Fold02.Rep10: gamma=0.75, lambda=0.75
## + Fold02.Rep10: gamma=1.00, lambda=0.75
## - Fold02.Rep10: gamma=1.00, lambda=0.75
## + Fold03.Rep10: gamma=0.25, lambda=0.75
## - Fold03.Rep10: gamma=0.25, lambda=0.75
## + Fold03.Rep10: gamma=0.50, lambda=0.75
## - Fold03.Rep10: gamma=0.50, lambda=0.75
## + Fold03.Rep10: gamma=0.75, lambda=0.75
## - Fold03.Rep10: gamma=0.75, lambda=0.75
## + Fold03.Rep10: gamma=1.00, lambda=0.75
## - Fold03.Rep10: gamma=1.00, lambda=0.75
## + Fold04.Rep10: gamma=0.25, lambda=0.75
## - Fold04.Rep10: gamma=0.25, lambda=0.75
## + Fold04.Rep10: gamma=0.50, lambda=0.75
## - Fold04.Rep10: gamma=0.50, lambda=0.75
## + Fold04.Rep10: gamma=0.75, lambda=0.75
## - Fold04.Rep10: gamma=0.75, lambda=0.75
## + Fold04.Rep10: gamma=1.00, lambda=0.75
## - Fold04.Rep10: gamma=1.00, lambda=0.75
## + Fold05.Rep10: gamma=0.25, lambda=0.75
## - Fold05.Rep10: gamma=0.25, lambda=0.75
## + Fold05.Rep10: gamma=0.50, lambda=0.75
## - Fold05.Rep10: gamma=0.50, lambda=0.75
## + Fold05.Rep10: gamma=0.75, lambda=0.75
## - Fold05.Rep10: gamma=0.75, lambda=0.75
## + Fold05.Rep10: gamma=1.00, lambda=0.75
## - Fold05.Rep10: gamma=1.00, lambda=0.75
## + Fold06.Rep10: gamma=0.25, lambda=0.75
## - Fold06.Rep10: gamma=0.25, lambda=0.75
## + Fold06.Rep10: gamma=0.50, lambda=0.75
## - Fold06.Rep10: gamma=0.50, lambda=0.75
## + Fold06.Rep10: gamma=0.75, lambda=0.75
## - Fold06.Rep10: gamma=0.75, lambda=0.75
## + Fold06.Rep10: gamma=1.00, lambda=0.75
## - Fold06.Rep10: gamma=1.00, lambda=0.75
## + Fold07.Rep10: gamma=0.25, lambda=0.75
## - Fold07.Rep10: gamma=0.25, lambda=0.75
## + Fold07.Rep10: gamma=0.50, lambda=0.75
## - Fold07.Rep10: gamma=0.50, lambda=0.75
## + Fold07.Rep10: gamma=0.75, lambda=0.75
## - Fold07.Rep10: gamma=0.75, lambda=0.75
## + Fold07.Rep10: gamma=1.00, lambda=0.75
## - Fold07.Rep10: gamma=1.00, lambda=0.75
## + Fold08.Rep10: gamma=0.25, lambda=0.75
## - Fold08.Rep10: gamma=0.25, lambda=0.75
## + Fold08.Rep10: gamma=0.50, lambda=0.75
## - Fold08.Rep10: gamma=0.50, lambda=0.75
## + Fold08.Rep10: gamma=0.75, lambda=0.75
## - Fold08.Rep10: gamma=0.75, lambda=0.75
## + Fold08.Rep10: gamma=1.00, lambda=0.75
## - Fold08.Rep10: gamma=1.00, lambda=0.75
## + Fold09.Rep10: gamma=0.25, lambda=0.75
## - Fold09.Rep10: gamma=0.25, lambda=0.75
## + Fold09.Rep10: gamma=0.50, lambda=0.75
## - Fold09.Rep10: gamma=0.50, lambda=0.75
## + Fold09.Rep10: gamma=0.75, lambda=0.75
## - Fold09.Rep10: gamma=0.75, lambda=0.75
## + Fold09.Rep10: gamma=1.00, lambda=0.75
## - Fold09.Rep10: gamma=1.00, lambda=0.75
## + Fold10.Rep10: gamma=0.25, lambda=0.75
## - Fold10.Rep10: gamma=0.25, lambda=0.75
## + Fold10.Rep10: gamma=0.50, lambda=0.75
## - Fold10.Rep10: gamma=0.50, lambda=0.75
## + Fold10.Rep10: gamma=0.75, lambda=0.75
## - Fold10.Rep10: gamma=0.75, lambda=0.75
## + Fold10.Rep10: gamma=1.00, lambda=0.75
## - Fold10.Rep10: gamma=1.00, lambda=0.75
## Aggregating results
## Selecting tuning parameters
## Fitting gamma = 0.25, lambda = 0.75 on full training setplot(rdafit)
rdafit## Regularized Discriminant Analysis
##
## 7720 samples
## 55 predictor
## 23 classes: 'ACC', 'BLCA', 'BRCA', 'CESC', 'COLORECTAL', 'DLBC', 'GLIOMA', 'HNSC', 'KIDNEY', 'LUNG', 'MESO', 'OV', 'PARALIVER', 'PCPG', 'PRAD', 'SARC', 'SKCM', 'STOPH', 'TGCT', 'THCA', 'THYM', 'UTERINE', 'UVM'
##
## No pre-processing
## Resampling: Cross-Validated (10 fold, repeated 10 times)
## Summary of sample sizes: 6949, 6948, 6944, 6950, 6948, 6947, ...
## Resampling results across tuning parameters:
##
## gamma logLoss AUC prAUC Accuracy Kappa Mean_F1
## 0.25 0.3736955 0.9971095 0.7359120 0.9409595 0.9358314 0.9280266
## 0.50 0.4191939 0.9966447 0.7683977 0.9321521 0.9262610 0.9209384
## 0.75 0.5667507 0.9955382 0.7724702 0.9124616 0.9048614 0.9025944
## 1.00 1.4541206 0.9865064 0.7275851 0.8686270 0.8574948 0.8583599
## Mean_Sensitivity Mean_Specificity Mean_Pos_Pred_Value Mean_Neg_Pred_Value
## 0.9334763 0.9973037 0.9296072 0.9972794
## 0.9261794 0.9968942 0.9236205 0.9968698
## 0.9079277 0.9959817 0.9083787 0.9959613
## 0.8708014 0.9939828 0.8593739 0.9938756
## Mean_Precision Mean_Recall Mean_Detection_Rate Mean_Balanced_Accuracy
## 0.9296072 0.9334763 0.04091128 0.9653900
## 0.9236205 0.9261794 0.04052835 0.9615368
## 0.9083787 0.9079277 0.03967224 0.9519547
## 0.8593739 0.8708014 0.03776639 0.9323921
##
## Tuning parameter 'lambda' was held constant at a value of 0.75
## Accuracy was used to select the optimal model using the largest value.
## The final values used for the model were gamma = 0.25 and lambda = 0.75.qual_col_pals = brewer.pal.info[brewer.pal.info$category == 'qual',]
col_vector = unlist(mapply(brewer.pal, qual_col_pals$maxcolors, rownames(qual_col_pals)))
pp2 <- preProcess(gt_all2[,-1], method = c("center", "nzv", "scale", "YeoJohnson"))
moddat <- predict(pp2, gt_all2)
pal = colorRampPalette(c("red","red","red", "black", "green","green","green"))(n = 128)
dd = dist(as.matrix(moddat[,-1]), method="maximum")
hh = hclust(dd, method="ward.D2")
moddat = moddat[order(as.numeric(moddat$mod_id)),]
# --- visualisation: heatmap ---#
pdf ('../../plots/heatmap_types.pdf', width=10, height=14)
heatmap.2(as.matrix(moddat[,-1]), trace="none", labRow = NA,
col=pal, #Rowv = NA,
RowSideColors = col_vector[as.numeric(factor(moddat$mod_id))],
hclustfun = function (x) {
hclust(dist(
x, method="maximum")
, method = "ward.D2")
}, key = F
)
dev.off()## quartz_off_screen
## 2# --- visualization: hierarchial clustering dendrogram ---#
set.seed(1234)
ss = sample(1:nrow(moddat), size=1000)
ss = moddat[ss,]
rownames(ss) = paste0(1:nrow(ss), ss$mod_id)
dd = dist(as.matrix(ss[,-1]), method="maximum")
hh = hclust(dd, method="ward.D2")
ColorDendrogram(hc=hh, y=col_vector[as.numeric(factor(ss$mod_id))],
branchlength = 6, main = "canberra"
#labels = ss$mod_id
)